CN111145173B - Plaque identification method, device, equipment and medium of coronary angiography image - Google Patents
Plaque identification method, device, equipment and medium of coronary angiography image Download PDFInfo
- Publication number
- CN111145173B CN111145173B CN201911423770.6A CN201911423770A CN111145173B CN 111145173 B CN111145173 B CN 111145173B CN 201911423770 A CN201911423770 A CN 201911423770A CN 111145173 B CN111145173 B CN 111145173B
- Authority
- CN
- China
- Prior art keywords
- coronary
- image
- angiography image
- coronary angiography
- plaque
- 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
- 238000002586 coronary angiography Methods 0.000 title claims abstract description 165
- 238000000034 method Methods 0.000 title claims abstract description 67
- 238000012549 training Methods 0.000 claims abstract description 49
- 210000004351 coronary vessel Anatomy 0.000 claims description 61
- 238000001514 detection method Methods 0.000 claims description 33
- 238000000605 extraction Methods 0.000 claims description 27
- 238000002583 angiography Methods 0.000 claims description 17
- 238000004590 computer program Methods 0.000 claims description 5
- 238000013528 artificial neural network Methods 0.000 description 8
- 201000000057 Coronary Stenosis Diseases 0.000 description 7
- 208000031481 Pathologic Constriction Diseases 0.000 description 7
- 210000004204 blood vessel Anatomy 0.000 description 7
- 230000036262 stenosis Effects 0.000 description 7
- 208000037804 stenosis Diseases 0.000 description 7
- 230000003287 optical effect Effects 0.000 description 6
- 206010011089 Coronary artery stenosis Diseases 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 238000010801 machine learning Methods 0.000 description 5
- 238000013135 deep learning Methods 0.000 description 4
- 230000002708 enhancing effect Effects 0.000 description 4
- 238000002372 labelling Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 2
- 239000013307 optical fiber Substances 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 206010002383 Angina Pectoris Diseases 0.000 description 1
- 238000012935 Averaging Methods 0.000 description 1
- 206010049418 Sudden Cardiac Death Diseases 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 210000000709 aorta Anatomy 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 208000010125 myocardial infarction Diseases 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
- 230000002792 vascular Effects 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
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30101—Blood vessel; Artery; Vein; Vascular
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the invention discloses a plaque identification method, a plaque identification device, plaque identification equipment and a plaque identification medium of a coronary angiography image, wherein the plaque identification method comprises the following steps: acquiring a coronary angiography image to be identified; and inputting the coronary angiography image to be identified into a complete plaque identification model to obtain an identification result output by the plaque identification model, wherein a training sample of the plaque identification model is generated based on the unlabeled coronary angiography image and the labeled coronary non-contrast image. According to the plaque identification method of the coronary angiography image, provided by the embodiment of the invention, the plaque identification model is trained based on the marked coronary non-contrast image and the unmarked coronary angiography image, so that the plaque identified by the plaque identification model is more accurate.
Description
Technical Field
The embodiment of the invention relates to the technical field of images, in particular to a plaque identification method, device, equipment and medium of a coronary angiography image.
Background
Coronary stenosis is a significant cause of angina pectoris, myocardial infarction and sudden cardiac death. Therefore, detection of coronary stenosis is particularly important. In the detection process of coronary artery stenosis, plaque extraction is required for the coronary image. The existing plaque detection is divided into hard plaque detection and soft plaque detection, and in the coronary artery extraction process, the contrast blood vessel, the bracket and the hard plaque also show higher CT values, so that the bracket and the hard plaque are extracted together during coronary artery extraction, and the soft plaque shows lower CT values and cannot be extracted. Thus, if a true coronary lumen is desired, the hard plaque and stent need to be removed from the coronary. However, the distribution area of the CT values of the hard plaque and a part of the stent overlap with the CT values of the contrast blood vessel, so that the threshold value for distinguishing the hard plaque from the stent and the CT values of the contrast blood vessel is difficult to set, and the extracted lumen accuracy is low.
Disclosure of Invention
The embodiment of the invention provides a plaque identification method, device, equipment and medium of a coronary angiography image, which are used for realizing the improvement of the plaque identification accuracy in the coronary angiography image and further improving the lumen extraction accuracy.
In a first aspect, an embodiment of the present invention provides a plaque identification method of a coronary angiography image, including:
Acquiring a coronary angiography image to be identified;
and inputting the coronary angiography image to be identified into a complete plaque identification model to obtain an identification result output by the plaque identification model, wherein a training sample of the plaque identification model is generated based on the unlabeled coronary angiography image and the labeled coronary non-contrast image.
In a second aspect, an embodiment of the present invention further provides a plaque identification apparatus for a coronary angiography image, including:
The contrast image acquisition module is used for acquiring a coronary contrast image to be identified;
and the image plaque recognition module is used for inputting the coronary angiography image to be recognized into a complete plaque recognition model to obtain a recognition result output by the plaque recognition model, wherein a training sample of the plaque recognition model is generated based on the unlabeled coronary angiography image and the labeled coronary non-contrast image.
In a third aspect, an embodiment of the present invention further provides a computer apparatus, the apparatus including:
One or more processors;
A storage means for storing one or more programs;
the one or more programs, when executed by the one or more processors, cause the one or more processors to implement a plaque identification method for a coronary angiography image as provided by any embodiment of the invention.
In a fourth aspect, embodiments of the present invention further provide a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a plaque identification method for a coronary angiographic image as provided by any embodiment of the present invention.
The embodiment of the invention obtains the coronary angiography image to be identified; and inputting the coronary angiography image to be identified into a fully trained plaque identification model to obtain an identification result output by the plaque identification model, wherein a training sample of the plaque identification model is generated based on the unlabeled coronary angiography image and the labeled coronary non-angiography image, and the plaque identification model is trained based on the labeled coronary non-angiography image and the unlabeled coronary angiography image, so that the plaque identified by the plaque identification model is more accurate.
Drawings
FIG. 1 is a flowchart of a method for plaque identification in a coronary angiographic image according to an embodiment of the present invention;
FIG. 2 is a flowchart of a plaque identification method for a coronary angiography image according to a second embodiment of the present invention;
Fig. 3 is a schematic structural diagram of a plaque recognition device of a coronary angiography image according to a third embodiment of the present invention;
fig. 4 is a schematic structural diagram of a computer device according to a fourth embodiment of the present invention.
Detailed Description
The invention is described in further detail below with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting thereof. It should be further noted that, for convenience of description, only some, but not all of the structures related to the present invention are shown in the drawings.
Example 1
Fig. 1 is a flowchart of a plaque identification method of a coronary angiography image according to an embodiment of the invention. The present embodiment is applicable to the case when recognizing plaque in a coronary angiographic image. The method may be performed by plaque recognition means of a coronary angiography image, which may be implemented in software and/or hardware, e.g. the plaque recognition means of the coronary angiography image may be configured in a computer device. As shown in fig. 1, the method includes:
S110, acquiring a coronary angiography image to be identified.
In this embodiment, the coronary angiography image to be identified may be an extracted coronary tree and a coronary centerline, and the coronary centerline is named. Optionally, after the coronary angiography image is acquired, seed points in the coronary angiography image are extracted, a coronary tree is extracted from the coronary angiography image based on the selected seed points according to methods such as a region growing algorithm or a level set algorithm, after the coronary extraction, a skeletonizing method or a centerline extraction method based on a level set is adopted to extract a coronary center tree from the coronary tree on the basis of a coronary mask, and then a model matching method or a deep learning method is used to name the extracted coronary center tree, so that the coronary angiography image to be identified is obtained.
S120, inputting the coronary angiography image to be identified into a complete plaque identification model to obtain an identification result output by the plaque identification model, wherein a training sample of the plaque identification model is generated based on the unlabeled coronary angiography image and the labeled coronary non-angiography image.
In order to solve the problems of inaccurate plaque identification and the like in the prior art, the embodiment of the invention identifies the coronary angiography image to be identified through a machine learning algorithm to obtain the identification result of the coronary angiography image to be identified. Specifically, the coronary angiography image to be identified is input into a fully trained plaque identification model, and an identification result output by the plaque identification model, namely, an identification result of the coronary angiography image to be identified is obtained. Alternatively, the plaque recognition model may be constructed based on a neural network. The neural network is a module built based on an artificial neural network (ARTIFICIAL NEURAL NETWORK, ANN). An artificial neural network is formed by interconnecting a large number of nodes (or neurons). Each node represents a specific output function, called the excitation function (activation function). Each connection between two nodes represents a weight, called a weight, for the signal passing through the connection. The neural network comprises a data input layer, an intermediate hiding layer and a data output layer. In this embodiment, the neural network may be a convolutional neural network (Convoltional Neural Networks, CNN), a generative network GENERATIVE ADVERSARIAL Networks, GAN, or other form of neural network model.
On the basis of the scheme, the plaque recognition model training method comprises the following steps:
acquiring a sample coronary angiography image and a marked sample coronary non-angiography image;
registering the sample coronary angiography image with the sample coronary non-contrast image to obtain a mark point of the sample coronary angiography image corresponding to the mark point in the sample coronary non-contrast image;
Generating an identification training sample based on the sample coronary angiography image and the mark points of the sample coronary angiography image, and training a pre-established plaque identification model by using the identification training sample to obtain a plaque identification model with complete training.
The contrast blood vessel in the contrast image is high in brightness, so that the CT value distribution areas of the hard plaque and the bracket in the contrast image are overlapped with the CT value of the contrast blood vessel, and the hard plaque and the bracket are difficult to distinguish. The blood vessel in the non-contrast image is low in brightness, so that the CT values of the hard plaque and the stent in the non-contrast image can be obviously distinguished from the CT values of the blood vessel. In this embodiment, by labeling the sample coronary non-contrast image, the labeling points of the sample coronary contrast image corresponding to the sample coronary non-contrast image are generated based on the labeling points in the sample coronary non-contrast image, and the labeled sample coronary contrast image is used to train the plaque recognition model established in advance, so that the plaque recognition model obtained by training is more accurate, and the recognition result obtained based on the plaque recognition model is more accurate.
Specifically, a sample coronary angiography image and a sample coronary non-angiography image corresponding to the sample coronary angiography image are obtained, hard plaques and stents in the sample coronary non-angiography image are manually marked to obtain a marked sample coronary non-angiography image, then the marked sample coronary non-angiography image and an unmarked sample coronary angiography image are subjected to image registration to obtain a sample coronary angiography image with marking points, and a pre-established plaque recognition model is trained by using the sample coronary angiography image with the marking points to obtain a trained complete plaque recognition model. The method for image registration of the labeled sample coronary non-contrast image and the unlabeled sample coronary contrast image is not limited herein.
For example, image registration may be performed using a gray-scale and template-based image registration method (e.g., a mean absolute difference algorithm, an absolute error sum algorithm, an error square sum algorithm, an average error square sum algorithm, a normalized product correlation algorithm, a sequential similarity detection algorithm, a hadamard transform algorithm, a local gray-scale value encoding algorithm, etc.), a feature-based image registration method (e.g., a point feature-based registration method, a line feature-based registration method, a region feature-based registration method, a local feature-based registration method, a global feature-based registration method, etc.), or a domain transform-based method (e.g., a phase correlation algorithm, a walsh transform method, etc.), or the like.
In the embodiment, the hard plaque and the stent in the sample coronary non-contrast image are manually marked, and the marking points in the sample coronary non-contrast image are determined according to the marking points in the sample coronary non-contrast image, so that the problem of inaccurate training sample caused by directly marking the sample coronary non-contrast image is solved, the sample accuracy of the plaque identification model is improved, and the identification accuracy of the plaque identification model is further improved.
The embodiment of the invention obtains the coronary angiography image to be identified; and inputting the coronary angiography image to be identified into a fully trained plaque identification model to obtain an identification result output by the plaque identification model, wherein a training sample of the plaque identification model is generated based on the unlabeled coronary angiography image and the labeled coronary non-angiography image, and the plaque identification model is trained based on the labeled coronary non-angiography image and the unlabeled coronary angiography image, so that the plaque identified by the plaque identification model is more accurate.
Example two
Fig. 2 is a flowchart of a plaque identification method of a coronary angiography image according to a second embodiment of the present invention. This embodiment is further optimized on the basis of the above-described embodiments. As shown in fig. 2, the method includes:
S210, acquiring a coronary angiography image, and extracting a coronary artery tree and a coronary artery central line from the coronary angiography image.
In this embodiment, extraction of the coronary artery tree is embodied. First, a coronary angiography image is acquired, and a coronary arterial tree and a coronary centerline are extracted from the coronary angiography image. Alternatively, the coronary artery tree may be extracted by a region growing method. Wherein the seed points may be one or more. Considering that the coronary tree is divided into a left coronary artery and a right coronary artery, more than two seed points are selected for extraction of the coronary tree. Alternatively, a plurality of seed points may be selected along the boundary line of the coronary artery. Alternatively, a template matching method may be used to extract seed points from the coronary angiographic image. By way of example, a template image of the marked coronary artery is prepared in advance, the acquired coronary angiography image and the template image are registered, the coronary position in the coronary angiography image can be determined by the coronary position in the template image, the opening position of the coronary artery on the aorta is determined, and the seed point is selected based on the opening position. However, the accuracy of the template matching method is low, and in order to improve the accuracy of the selected seed points, a deep learning method can be used for extracting the seed points.
In one embodiment of the present invention, the extracting coronary artery tree from the coronary angiographic image includes:
inputting the coronary angiography image into a fully trained seed point extraction model to obtain seed points output by the seed point extraction model; performing image enhancement on the coronary angiography image to obtain a coronary angiography enhanced image of the coronary angiography image; the coronary artery tree is extracted from the coronary angiography-enhanced image based on the seed points by a region growing algorithm.
In this embodiment, the seed point extraction model built in advance is trained by a deep learning method by marking the positions (such as the coronary artery openings) of seed points in a large number of sample coronary angiography images, so as to obtain a fully trained seed point extraction model. After the coronary angiography image is obtained, the coronary angiography image is input into a fully trained seed point extraction model, so that the seed points output by the seed point extraction model are obtained and used as the seed points of the coronary angiography image.
In order to make the extraction of the coronary artery tree more accurate, the coronary angiography image is enhanced before the extraction of the coronary artery tree is performed, and a coronary angiography enhanced image is obtained. The method of enhancing the coronary angiographic image is not limited herein. Illustratively, a line enhancement method of a black plug (Hessian) matrix may be used to obtain the coronary angiography enhanced image, and a depth learning-based method may also be used to obtain the coronary angiography enhanced image. Alternatively, obtaining the coronary angiography enhanced image based on the deep learning method may be: and inputting the coronary angiography image into a fully trained enhancement model to obtain a coronary angiography enhancement image output by the enhancement model. The enhancement model is obtained by training based on a sample coronary angiography image and a sample coronary angiography enhancement image corresponding to the sample coronary angiography image.
In the present embodiment, the order of the operations of enhancing the coronary angiographic image and extracting the seed points is not limited. I.e. seed points can be extracted from the coronary angiography image firstly, and then the coronary angiography image is enhanced; or enhancing the coronary angiography image, and extracting seed points from the coronary angiography image; the operations of enhancing the coronary angiographic image and extracting the seed points from the coronary angiographic image may also be performed simultaneously.
After obtaining the seed point of the coronary angiography image and obtaining the coronary angiography enhancement image, a certain enhancement threshold value can be set in the coronary angiography enhancement image, region growth is carried out from the position of the seed point, and the extracted coronary arterial tree is obtained after the region growth is completed. The enhanced image can be combined with the original image, namely, a certain gray threshold is set in the coronary angiography image, a certain enhanced threshold is set in the coronary angiography enhanced image, region growth is carried out from the position of the seed point, and the extracted coronary arterial tree is obtained after the region growth is completed. After the coronary artery tree is extracted, a skeletonizing method or a centerline extraction method based on a level set can be adopted to extract the coronary artery centerline.
And S220, naming the coronary artery central line to obtain the coronary artery contrast image to be identified.
In this embodiment, after the coronary artery centerline is extracted, the coronary artery centerline is named so that the detection result is easy to locate in the subsequent plaque identification or coronary artery stenosis detection. On the basis of the coronary artery centerline tree, each coronary artery centerline can be named, and the main coronary artery centerline names comprise a left coronary artery trunk, a left anterior descending branch, a left circumflex branch, a left diagonal branch, a left blunt edge branch, a right coronary artery, a right descending branch and the like. Alternatively, coronary centerline naming may employ a model matching approach. Illustratively, according to sample data, manually marking the names of coronary artery central lines, averaging all central lines on the sample, making a central line average model, matching the model to an image where the coronary artery central lines are located by using a point registration method and the like, and naming the name of each coronary artery central line as the name of the central line on the model similar to the coronary artery central line. In order to reduce the error rate, besides model matching, a machine learning method can be adopted, so that the accuracy rate is further improved.
In one embodiment of the present invention, the naming the coronary centerline includes:
And inputting the coronary artery central line into a central line naming model which is completely trained, and obtaining a naming result output by the central line naming model.
Preferably, the coronary artery center line is named by using a machine learning method, and the coronary artery center line is named by using the machine learning method, so that the naming of the coronary artery center line is more accurate. Specifically, the coronary artery central line is input into a central line naming model which is completely trained, a naming result output by the central line naming model is obtained, and the coronary artery central line is named according to the output naming result.
On the basis of the scheme, the training method of the central line naming model comprises the following steps:
Acquiring a central line parameter in a central line model and a central line name corresponding to the central line model;
Generating a named training sample based on the central line parameters and the central line names corresponding to the central line models, and training the pre-established central line named models by using the named training sample to obtain the central line named models with complete training.
Optionally, the central line parameter may be a characteristic parameter of each coronary artery central line in the coronary artery central line model, such as a length of the coronary artery central line, an angle of the coronary artery central line, and the like, and the central line parameter and the central line model are used to train the pre-established central line naming model, so as to obtain a central line naming model with complete training.
S230, inputting the coronary angiography image to be identified into a complete plaque identification model to obtain an identification result output by the plaque identification model, wherein a training sample of the plaque identification model is generated based on the unlabeled coronary angiography image and the labeled coronary non-angiography image.
S240, generating a coronary angiography image to be detected based on the coronary angiography image to be identified and the identification result.
In this embodiment, after identifying the plaque and the stent in the coronary angiography image to be identified, the plaque and the stent in the coronary angiography image to be identified are removed, so as to obtain the coronary angiography image to be detected which only includes the vascular lumen.
S250, carrying out narrow detection on the coronary angiography image to be detected, and outputting a detection result.
And after obtaining the coronary angiography image to be detected, carrying out narrow detection on the coronary angiography image to be detected. Alternatively, the cross-sectional profile, the area and the minor diameter (or the diameter) of each coronary artery can be calculated, fitting is performed, and whether the coronary artery is narrow or not is judged according to the fitted result. Alternatively, a stenosis is considered to exist here if the diameter of the coronary vessel at a location is smaller than the diameter of the vessel upstream and downstream of it and a proportional threshold is reached.
In another embodiment of the present invention, the stenosis detection of the coronary angiographic image to be detected may also be performed by means of machine learning. Optionally, the coronary angiography image to be detected is input into a fully trained stenosis detection model, and a detection result output by the stenosis detection model is obtained. The training complete stenosis detection model is obtained by training based on a sample coronary angiography image and a detection result corresponding to the sample coronary angiography image.
Based on the scheme, after the narrow detection is carried out on the tube, the detection result can be output. In this embodiment, operations of coronary artery segmentation, center line extraction, center line naming, plaque and stent detection and stenosis detection are automatically completed, and results corresponding to the above steps can be displayed on the same interface, so that when coronary artery stenosis detection is performed, the user can edit inaccurate places in the detection process without switching the interface or dividing the interface. After receiving the editing operation of the user, the result of the subsequent affected step is automatically updated according to the editing operation of the user.
According to the technical scheme provided by the embodiment of the invention, the accuracy of coronary artery extraction, center line naming, plaque detection and stenosis detection is improved, so that the detection result can be obtained without user intervention in the coronary artery stenosis detection process, and the results of all the steps are directly presented to the user on the same interface, so that the technical effect of directly modifying the final detection result is achieved without the need of the user to check and modify in steps under the condition that the automatic extraction cannot meet the requirement or the user needs to modify part of parameters.
Example III
Fig. 3 is a schematic structural diagram of a plaque identification device for coronary angiography image according to a third embodiment of the present invention. The plaque recognition means of the coronary angiography image may be implemented in software and/or hardware, e.g. the plaque recognition means of the coronary angiography image may be configured in a computer device. As shown in fig. 3, the apparatus includes a contrast image acquisition module 310 and an image plaque identification module 320, wherein:
a contrast image acquisition module 310, configured to acquire a coronary contrast image to be identified;
The image plaque recognition module 320 is configured to input the coronary angiography image to be recognized into a trained complete plaque recognition model, and obtain a recognition result output by the plaque recognition model, where a training sample of the plaque recognition model is generated based on an unlabeled coronary angiography image and a labeled coronary non-contrast image.
According to the embodiment of the invention, the coronary angiography image to be identified is acquired through the angiography image acquisition module; the image plaque recognition module inputs the coronary angiography image to be recognized into a complete plaque recognition model to obtain a recognition result output by the plaque recognition model, wherein a training sample of the plaque recognition model is generated based on the unlabeled coronary angiography image and the labeled coronary non-angiography image, and the plaque recognition model is trained based on the labeled coronary non-angiography image and the unlabeled coronary angiography image, so that the plaque recognized by the plaque recognition model is more accurate.
On the basis of the scheme, the device further comprises an identification model training module for:
acquiring a sample coronary angiography image and a marked sample coronary non-angiography image;
registering the sample coronary angiography image with the sample coronary non-contrast image to obtain a mark point of the sample coronary angiography image corresponding to the mark point in the sample coronary non-contrast image;
Generating an identification training sample based on the sample coronary angiography image and the mark points of the sample coronary angiography image, and training a pre-established plaque identification model by using the identification training sample to obtain a plaque identification model with complete training.
On the basis of the above-mentioned scheme, the contrast image acquisition module 310 includes:
a coronary artery extraction unit for acquiring a coronary angiography image, and extracting a coronary artery tree and a coronary artery central line from the coronary angiography image;
And the central line naming unit is used for naming the central line of the coronary artery to obtain the coronary angiography image to be identified.
On the basis of the above scheme, the coronary artery extraction unit is specifically configured to:
Inputting the coronary angiography image into a fully trained seed point extraction model to obtain seed points output by the seed point extraction model;
Performing image enhancement on the coronary angiography image to obtain a coronary angiography enhanced image of the coronary angiography image;
The coronary artery tree is extracted from the coronary angiography-enhanced image based on the seed points by a region growing algorithm.
Based on the scheme, the central line naming unit is specifically used for:
And inputting the coronary artery central line into a central line naming model which is completely trained, and obtaining a naming result output by the central line naming model.
Based on the scheme, the device further comprises a naming model training module for:
Acquiring a central line parameter in a central line model and a central line name corresponding to the central line model;
Generating a named training sample based on the central line parameters and the central line names corresponding to the central line models, and training the pre-established central line named models by using the named training sample to obtain the central line named models with complete training.
On the basis of the scheme, the device further comprises a coronary artery stenosis detection module for:
generating a coronary angiography image to be detected based on the coronary angiography image to be identified and the identification result;
and carrying out narrow detection on the coronary angiography image to be detected, and outputting a detection result.
The plaque recognition device of the coronary angiography image provided by the embodiment of the invention can execute the plaque recognition method of the coronary angiography image provided by any embodiment, and has the corresponding functional modules and beneficial effects of the execution method.
Example IV
Fig. 4 is a schematic structural diagram of a computer device according to a fourth embodiment of the present invention. Fig. 4 illustrates a block diagram of an exemplary computer device 412 suitable for use in implementing embodiments of the invention. The computer device 412 shown in fig. 4 is only an example and should not be construed as limiting the functionality and scope of use of embodiments of the invention.
As shown in FIG. 4, computer device 412 is in the form of a general purpose computing device. Components of computer device 412 may include, but are not limited to: one or more processors 416, a system memory 428, and a bus 418 that connects the various system components (including the system memory 428 and the processors 416).
Bus 418 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor 416, or a local bus using any of a variety of bus architectures. By way of example, and not limitation, such architectures include Industry Standard Architecture (ISA) bus, micro channel architecture (MAC) bus, enhanced ISA bus, video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus.
Computer device 412 typically includes a variety of computer system readable media. Such media can be any available media that is accessible by computer device 412 and includes both volatile and nonvolatile media, removable and non-removable media.
The system memory 428 may include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM) 430 and/or cache memory 432. The computer device 412 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage 434 may be used to read from or write to non-removable, non-volatile magnetic media (not shown in FIG. 4, commonly referred to as a "hard disk drive"). Although not shown in fig. 4, a magnetic disk drive for reading from and writing to a removable non-volatile magnetic disk (e.g., a "floppy disk"), and an optical disk drive for reading from or writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In such cases, each drive may be coupled to bus 418 via one or more data medium interfaces. Memory 428 may include at least one program product having a set (e.g., at least one) of program modules configured to carry out the functions of embodiments of the invention.
A program/utility 440 having a set (at least one) of program modules 442 may be stored in, for example, memory 428, such program modules 442 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each or some combination of which may include an implementation of a network environment. Program modules 442 generally perform the functions and/or methodologies in the described embodiments of the invention.
The computer device 412 may also communicate with one or more external devices 414 (e.g., keyboard, pointing device, display 424, etc.), one or more devices that enable a user to interact with the computer device 412, and/or any devices (e.g., network card, modem, etc.) that enable the computer device 412 to communicate with one or more other computing devices. Such communication may occur through an input/output (I/O) interface 422. Moreover, computer device 412 may also communicate with one or more networks such as a Local Area Network (LAN), a Wide Area Network (WAN) and/or a public network, such as the Internet, through network adapter 420. As shown, network adapter 420 communicates with other modules of computer device 412 over bus 418. It should be appreciated that although not shown, other hardware and/or software modules may be used in connection with computer device 412, including, but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, data backup storage systems, and the like.
The processor 416 executes various functional applications and data processing by running programs stored in the system memory 428, for example, to implement a plaque identification method for a coronary angiographic image provided by an embodiment of the present invention, the method comprising:
Acquiring a coronary angiography image to be identified;
and inputting the coronary angiography image to be identified into a complete plaque identification model to obtain an identification result output by the plaque identification model, wherein a training sample of the plaque identification model is generated based on the unlabeled coronary angiography image and the labeled coronary non-contrast image.
Of course, those skilled in the art will understand that the processor may also implement the technical solution of the plaque identification method of the coronary angiography image provided in any embodiment of the present invention.
Example five
The fifth embodiment of the present invention further provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a plaque identification method for a coronary angiography image as provided by the embodiment of the present invention, the method comprising:
Acquiring a coronary angiography image to be identified;
and inputting the coronary angiography image to be identified into a complete plaque identification model to obtain an identification result output by the plaque identification model, wherein a training sample of the plaque identification model is generated based on the unlabeled coronary angiography image and the labeled coronary non-contrast image.
Of course, the computer-readable storage medium provided by the embodiments of the present invention, on which the computer program stored, is not limited to the method operations described above, but may also perform the related operations in the plaque identification method of the coronary angiographic image provided by any embodiment of the present invention.
The computer storage media of embodiments of the invention may take the form of any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
The computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, either in baseband or as part of a carrier wave. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Computer program code for carrying out operations of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, smalltalk, C ++ and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider).
Note that the above is only a preferred embodiment of the present invention and the technical principle applied. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, while the invention has been described in connection with the above embodiments, the invention is not limited to the embodiments, but may be embodied in many other equivalent forms without departing from the spirit or scope of the invention, which is set forth in the following claims.
Claims (9)
1. A method for plaque identification in a coronary angiographic image, comprising:
Acquiring a coronary angiography image to be identified;
Inputting the coronary angiography image to be identified into a fully trained plaque identification model to obtain an identification result output by the plaque identification model, wherein a training sample of the plaque identification model is generated based on an unlabeled coronary angiography image and a labeled coronary non-angiography image;
wherein the acquiring the coronary angiography image to be identified comprises:
Acquiring a coronary angiography image, and extracting a coronary artery tree and a coronary artery central line from the coronary angiography image;
And naming the coronary artery central line to obtain the coronary artery contrast image to be identified.
2. The method of claim 1, wherein the training method of the plaque recognition model comprises:
acquiring a sample coronary angiography image and a marked sample coronary non-angiography image;
registering the sample coronary angiography image with the sample coronary non-contrast image to obtain a mark point of the sample coronary angiography image corresponding to the mark point in the sample coronary non-contrast image;
Generating an identification training sample based on the sample coronary angiography image and the mark points of the sample coronary angiography image, and training a pre-established plaque identification model by using the identification training sample to obtain a plaque identification model with complete training.
3. The method of claim 1, wherein the extracting the coronary artery tree from the coronary angiography image comprises:
Inputting the coronary angiography image into a fully trained seed point extraction model to obtain seed points output by the seed point extraction model;
Performing image enhancement on the coronary angiography image to obtain a coronary angiography enhanced image of the coronary angiography image;
The coronary artery tree is extracted from the coronary angiography-enhanced image based on the seed points by a region growing algorithm.
4. The method of claim 1, wherein naming the coronary centerline comprises:
And inputting the coronary artery central line into a central line naming model which is completely trained, and obtaining a naming result output by the central line naming model.
5. The method of claim 4, wherein the training method of the centerline naming model comprises:
Acquiring a central line parameter in a central line model and a central line name corresponding to the central line model;
Generating a named training sample based on the central line parameters and the central line names corresponding to the central line models, and training the pre-established central line named models by using the named training sample to obtain the central line named models with complete training.
6. The method of claim 1, further comprising, after obtaining the identification result output by the plaque identification model:
generating a coronary angiography image to be detected based on the coronary angiography image to be identified and the identification result;
and carrying out narrow detection on the coronary angiography image to be detected, and outputting a detection result.
7. A plaque identification device for a coronary angiography image, comprising:
The contrast image acquisition module is used for acquiring a coronary contrast image to be identified;
The image plaque recognition module is used for inputting the coronary angiography image to be recognized into a complete plaque recognition model to obtain a recognition result output by the plaque recognition model, wherein a training sample of the plaque recognition model is generated based on an unlabeled coronary angiography image and a labeled coronary non-contrast image;
Wherein, the contrast image acquisition module includes:
a coronary artery extraction unit for acquiring a coronary angiography image, and extracting a coronary artery tree and a coronary artery central line from the coronary angiography image;
And the central line naming unit is used for naming the central line of the coronary artery to obtain the coronary angiography image to be identified.
8. A computer device, the device comprising:
One or more processors;
A storage means for storing one or more programs;
The one or more programs, when executed by the one or more processors, cause the one or more processors to implement the plaque identification method of a coronary angiography image according to any one of claims 1-6.
9. A computer-readable storage medium, on which a computer program is stored, characterized in that the program, when being executed by a processor, implements a plaque identification method of a coronary angiography image according to any one of claims 1-6.
Priority Applications (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911423770.6A CN111145173B (en) | 2019-12-31 | 2019-12-31 | Plaque identification method, device, equipment and medium of coronary angiography image |
| EP20910740.8A EP4066207A4 (en) | 2019-12-31 | 2020-12-29 | Systems and methods for image processing |
| PCT/CN2020/141089 WO2021136304A1 (en) | 2019-12-31 | 2020-12-29 | Systems and methods for image processing |
| US17/810,322 US20220335613A1 (en) | 2019-12-31 | 2022-06-30 | Systems and methods for image processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911423770.6A CN111145173B (en) | 2019-12-31 | 2019-12-31 | Plaque identification method, device, equipment and medium of coronary angiography image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111145173A CN111145173A (en) | 2020-05-12 |
| CN111145173B true CN111145173B (en) | 2024-04-26 |
Family
ID=70523018
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911423770.6A Active CN111145173B (en) | 2019-12-31 | 2019-12-31 | Plaque identification method, device, equipment and medium of coronary angiography image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111145173B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP4066207A4 (en) * | 2019-12-31 | 2023-01-25 | Shanghai United Imaging Healthcare Co., Ltd. | Systems and methods for image processing |
| CN111681226A (en) * | 2020-06-09 | 2020-09-18 | 上海联影医疗科技有限公司 | Target tissue positioning method and device based on blood vessel identification |
| CN111862038B (en) * | 2020-07-17 | 2024-05-14 | 中国医学科学院阜外医院 | Plaque detection method, plaque detection device, plaque detection equipment and plaque detection medium |
| CN112017167A (en) * | 2020-08-24 | 2020-12-01 | 杭州深睿博联科技有限公司 | Coronary artery central line generation method and device based on bidirectional coronary artery blood vessel tracking |
| CN112419308B (en) * | 2020-12-07 | 2023-06-23 | 上海联影医疗科技股份有限公司 | Plaque evaluation method, plaque evaluation device, electronic equipment and storage medium |
| CN112700421B (en) * | 2021-01-04 | 2022-03-25 | 推想医疗科技股份有限公司 | Coronary image classification method and device |
| WO2022228463A1 (en) * | 2021-04-28 | 2022-11-03 | 上海博动医疗科技股份有限公司 | Coronary artery plaque state assessment method and apparatus, and electronic device |
| CN113222964B (en) * | 2021-05-27 | 2021-11-12 | 推想医疗科技股份有限公司 | Method and device for generating coronary artery central line extraction model |
| CN113436177A (en) * | 2021-07-01 | 2021-09-24 | 万里云医疗信息科技（北京）有限公司 | Image processing method and device, electronic equipment and computer readable storage medium |
| CN113628193B (en) * | 2021-08-12 | 2022-07-15 | 推想医疗科技股份有限公司 | Method, device and system for determining blood vessel stenosis rate and storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2405260A1 (en) * | 2010-07-09 | 2012-01-11 | Alta Lab S.r.l. | Method and apparatus for performing non-invasive x-ray inspections of objects |
| CN102945328A (en) * | 2012-11-22 | 2013-02-27 | 北京理工大学 | Simulation method for X ray angiography image based on graphic processing unit (GPU) parallel computation |
| CN107871318A (en) * | 2017-11-16 | 2018-04-03 | 吉林大学 | A kind of coronary calcification plaque detection method based on model migration |
| CN108171702A (en) * | 2018-01-18 | 2018-06-15 | 平安科技（深圳）有限公司 | Vulnerable plaque recognition methods, application server and computer readable storage medium |
| CN108805923A (en) * | 2018-06-01 | 2018-11-13 | 沈阳东软医疗***有限公司 | Physiological parameter acquisition methods and device |
| CN109214238A (en) * | 2017-06-30 | 2019-01-15 | 百度在线网络技术（北京）有限公司 | Multi-object tracking method, device, equipment and storage medium |
| JP6491391B1 (en) * | 2018-11-22 | 2019-03-27 | 国立大学法人信州大学 | Blood flow dynamics analysis system, blood flow dynamics analysis method, and program |
| CN110310271A (en) * | 2019-07-01 | 2019-10-08 | 无锡祥生医疗科技股份有限公司 | Property method of discrimination, storage medium and the Vltrasonic device of carotid plaques |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9053551B2 (en) * | 2012-05-23 | 2015-06-09 | International Business Machines Corporation | Vessel identification using shape and motion mapping for coronary angiogram sequences |
| US9805463B2 (en) * | 2013-08-27 | 2017-10-31 | Heartflow, Inc. | Systems and methods for predicting location, onset, and/or change of coronary lesions |
| US9642586B2 (en) * | 2014-08-18 | 2017-05-09 | Siemens Aktiengesellschaft | Computer-aided analysis of medical images |
-
2019
- 2019-12-31 CN CN201911423770.6A patent/CN111145173B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2405260A1 (en) * | 2010-07-09 | 2012-01-11 | Alta Lab S.r.l. | Method and apparatus for performing non-invasive x-ray inspections of objects |
| CN102945328A (en) * | 2012-11-22 | 2013-02-27 | 北京理工大学 | Simulation method for X ray angiography image based on graphic processing unit (GPU) parallel computation |
| CN109214238A (en) * | 2017-06-30 | 2019-01-15 | 百度在线网络技术（北京）有限公司 | Multi-object tracking method, device, equipment and storage medium |
| CN107871318A (en) * | 2017-11-16 | 2018-04-03 | 吉林大学 | A kind of coronary calcification plaque detection method based on model migration |
| CN108171702A (en) * | 2018-01-18 | 2018-06-15 | 平安科技（深圳）有限公司 | Vulnerable plaque recognition methods, application server and computer readable storage medium |
| WO2019140857A1 (en) * | 2018-01-18 | 2019-07-25 | 平安科技（深圳）有限公司 | Vulnerable plaque identification method, application server, and computer readable storage medium |
| CN108805923A (en) * | 2018-06-01 | 2018-11-13 | 沈阳东软医疗***有限公司 | Physiological parameter acquisition methods and device |
| JP6491391B1 (en) * | 2018-11-22 | 2019-03-27 | 国立大学法人信州大学 | Blood flow dynamics analysis system, blood flow dynamics analysis method, and program |
| CN110310271A (en) * | 2019-07-01 | 2019-10-08 | 无锡祥生医疗科技股份有限公司 | Property method of discrimination, storage medium and the Vltrasonic device of carotid plaques |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111145173A (en) | 2020-05-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111145173B (en) | Plaque identification method, device, equipment and medium of coronary angiography image | |
| US10115039B2 (en) | Method and system for machine learning based classification of vascular branches | |
| Chen et al. | Curve-like structure extraction using minimal path propagation with backtracking | |
| CN108805134B (en) | Construction method and application of aortic dissection model | |
| CN111696089A (en) | Arteriovenous determining method, device, equipment and storage medium | |
| US8612890B2 (en) | Labeling a segmented object | |
| CN110298844B (en) | X-ray radiography image blood vessel segmentation and identification method and device | |
| CN109478321B (en) | Automated 3D brain tumor segmentation and classification | |
| CN109087306A (en) | Arteries iconic model training method, dividing method, device and electronic equipment | |
| US10548552B2 (en) | Method and device for generating anatomical labels for a physiological tree structure | |
| US8805044B2 (en) | Segmenting pulmonary arteries | |
| CN111145160B (en) | Method, device, server and medium for determining coronary artery branches where calcified regions are located | |
| CN110110723B (en) | Method and device for automatically extracting target area in image | |
| CN110176013B (en) | Blood vessel segmentation method, blood vessel segmentation device, electronic device and storage medium | |
| CN110706272B (en) | Blood vessel centerline extraction method and device, electronic equipment and storage medium | |
| CN111640124B (en) | Blood vessel extraction method, device, equipment and storage medium | |
| US20220366679A1 (en) | Methods and systems for training learning network for medical image analysis | |
| US20240078676A1 (en) | Interactive coronary labeling using interventional x-ray images and deep learning | |
| CN113889238B (en) | Image identification method and device, electronic equipment and storage medium | |
| CN112801999B (en) | Method and device for determining heart coronary artery dominance | |
| CN111127400A (en) | Method and device for detecting breast lesions | |
| CN117373070A (en) | Method and device for labeling blood vessel segments, electronic equipment and storage medium | |
| Xiao et al. | Automatic vasculature identification in coronary angiograms by adaptive geometrical tracking | |
| CN112614143A (en) | Image segmentation method and device, electronic equipment and storage medium | |
| CN115908418A (en) | Method, system, equipment and medium for determining central line of aorta CT image |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 201807 Shanghai City, north of the city of Jiading District Road No. 2258 Applicant after: Shanghai Lianying Medical Technology Co.,Ltd. Address before: 201807 Shanghai City, north of the city of Jiading District Road No. 2258 Applicant before: SHANGHAI UNITED IMAGING HEALTHCARE Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |