CN114469174A - Artery plaque identification method and system based on ultrasonic scanning video - Google Patents
Artery plaque identification method and system based on ultrasonic scanning video Download PDFInfo
- Publication number
- CN114469174A CN114469174A CN202111556641.1A CN202111556641A CN114469174A CN 114469174 A CN114469174 A CN 114469174A CN 202111556641 A CN202111556641 A CN 202111556641A CN 114469174 A CN114469174 A CN 114469174A
- Authority
- CN
- China
- Prior art keywords
- identification
- image
- ultrasonic scanning
- arterial plaque
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 30
- 210000001367 artery Anatomy 0.000 title claims abstract description 21
- 238000012545 processing Methods 0.000 claims abstract description 13
- 238000002604 ultrasonography Methods 0.000 claims description 63
- 238000000605 extraction Methods 0.000 claims description 24
- 238000011176 pooling Methods 0.000 claims description 23
- 238000013528 artificial neural network Methods 0.000 claims description 7
- 230000004927 fusion Effects 0.000 claims description 6
- 230000000306 recurrent effect Effects 0.000 claims description 6
- 230000017105 transposition Effects 0.000 claims description 2
- 238000001514 detection method Methods 0.000 abstract description 2
- 210000004204 blood vessel Anatomy 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000036770 blood supply Effects 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 208000024891 symptom Diseases 0.000 description 2
- 206010002383 Angina Pectoris Diseases 0.000 description 1
- 206010003210 Arteriosclerosis Diseases 0.000 description 1
- 206010003211 Arteriosclerosis coronary artery Diseases 0.000 description 1
- 208000031481 Pathologic Constriction Diseases 0.000 description 1
- 210000003484 anatomy Anatomy 0.000 description 1
- 208000011775 arteriosclerosis disease Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000017531 blood circulation Effects 0.000 description 1
- 230000002490 cerebral effect Effects 0.000 description 1
- 206010008118 cerebral infarction Diseases 0.000 description 1
- 208000026106 cerebrovascular disease Diseases 0.000 description 1
- 238000004195 computer-aided diagnosis Methods 0.000 description 1
- 208000029078 coronary artery disease Diseases 0.000 description 1
- 208000026758 coronary atherosclerosis Diseases 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 150000002632 lipids Chemical class 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000003211 malignant effect Effects 0.000 description 1
- 208000010125 myocardial infarction Diseases 0.000 description 1
- 208000031225 myocardial ischemia Diseases 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000036262 stenosis Effects 0.000 description 1
- 208000037804 stenosis Diseases 0.000 description 1
- 230000006441 vascular event Effects 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/08—Detecting organic movements or changes, e.g. tumours, cysts, swellings
- A61B8/0891—Detecting organic movements or changes, e.g. tumours, cysts, swellings for diagnosis of blood vessels
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/08—Detecting organic movements or changes, e.g. tumours, cysts, swellings
- A61B8/0833—Detecting organic movements or changes, e.g. tumours, cysts, swellings involving detecting or locating foreign bodies or organic structures
- A61B8/085—Detecting organic movements or changes, e.g. tumours, cysts, swellings involving detecting or locating foreign bodies or organic structures for locating body or organic structures, e.g. tumours, calculi, blood vessels, nodules
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/52—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/5207—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves involving processing of raw data to produce diagnostic data, e.g. for generating an image
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/52—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/5215—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves involving processing of medical diagnostic data
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/52—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/5215—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves involving processing of medical diagnostic data
- A61B8/5223—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves involving processing of medical diagnostic data for extracting a diagnostic or physiological parameter from medical diagnostic data
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Heart & Thoracic Surgery (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Pathology (AREA)
- Radiology & Medical Imaging (AREA)
- Biomedical Technology (AREA)
- Veterinary Medicine (AREA)
- Medical Informatics (AREA)
- Physics & Mathematics (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Vascular Medicine (AREA)
- Physiology (AREA)
- Ultra Sonic Daignosis Equipment (AREA)
Abstract
The invention provides an artery plaque identification method and system based on an ultrasonic scanning video, which relate to the technical field of medical image processing and comprise the following steps: acquiring an ultrasonic scanning video, and extracting continuous multi-frame ultrasonic scanning images from the ultrasonic scanning video; respectively carrying out arterial plaque identification on each ultrasonic scanning image to obtain a corresponding image identification result; and processing each image identification result to obtain an artery plaque identification result of the ultrasonic scanning video. The method has the advantages that the artery plaque identification is carried out through the continuous multi-frame ultrasonic scanning images in the ultrasonic scanning video, the dynamic image information of the continuous frames can be captured, further, the artery plaque with the missing and the artery plaque which is difficult to identify by a single ultrasonic scanning image can be well detected, the identification rate is improved, and the false positive detection rate is reduced.
Description
Technical Field
The invention relates to the technical field of medical image processing, in particular to an arterial plaque identification method and system based on an ultrasonic scanning video.
Background
Arterial plaque refers to the occurrence of arteriosclerosis, a deposit of lipids, within the wall of a blood vessel at a certain site of an artery. With the time extension, the arterial plaque will grow slowly in the artery, resulting in the gradual narrowing of the lumen of the blood vessel, and then the symptom of insufficient blood supply will appear, the carotid plaque will cause the cerebral blood supply insufficiency, the coronary atherosclerosis will cause the myocardial ischemia, and cause the angina pectoris, more seriously, as the plaque grows larger, the plaque will break, then flow to all parts of the body along our blood vessel, and then flow to where, and then block where, once block, we will appear the clinical symptoms, such as the malignant vascular events of myocardial infarction, cerebral infarction. This is also a serious consequence of plaque.
Due to low price and safety, the ultrasonic imaging is the first choice method for screening the arterial plaque in clinic at present, can reflect the anatomical structure and plaque characteristics of the blood vessel, and can provide indexes such as blood flow velocity in the blood vessel, stenosis rate of the blood vessel and the like. Due to the complexity of the medical images themselves, plaque identification based on ultrasound images requires review by an experienced physician. With the development of computer-aided diagnosis technology, plaque identification on an ultrasonic image by adopting a machine learning mode currently appears.
At present, the identification of the arterial plaque through the ultrasonic image is mainly based on a single ultrasonic image processing mode, some traditional image segmentation identification algorithms are used, some modes are based on deep learning, and the two-dimensional segmentation and identification network is mainly focused. Because the ultrasound image itself has many noises and artifacts, even a doctor needs to repeatedly check the video to determine the images in the process of diagnosis, the processing mode based on a single image does not consider the difference between different frames, cannot capture dynamic information, causes low recognition rate and high false positive, causes misdiagnosis or missed diagnosis, and is difficult or impossible to meet the needs.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides an arterial plaque identification method based on an ultrasonic scanning video, which comprises the following steps:
step S1, acquiring an ultrasonic scanning video, and extracting continuous multi-frame ultrasonic scanning images from the ultrasonic scanning video;
step S2, performing arterial plaque identification on each ultrasonic scanning image to obtain corresponding image identification results;
and step S3, processing each image recognition result to obtain an artery plaque recognition result of the ultrasonic scanning video.
Preferably, the step S2 includes:
step S21, respectively carrying out feature extraction on each ultrasonic scanning image to obtain an interested area in each ultrasonic scanning image;
step S22, performing arterial plaque identification on each region of interest to obtain plaque position information in each region of interest as the image identification result.
Preferably, before the step S21, the method further includes dividing each of the ultrasound scanning images into a plurality of groups;
in step S21, feature extraction is performed on each group of ultrasound scanning images to obtain the region of interest corresponding to each group of ultrasound scanning images;
in the step S22, arterial plaque identification is performed on the region of interest corresponding to each group of the ultrasound scanning images, respectively, to obtain the image identification result corresponding to each group of the ultrasound scanning images.
Preferably, in step S21, a global attention network is adopted to perform feature extraction on each group of the ultrasound scanning images respectively to obtain the region of interest in each group of the ultrasound scanning images.
Preferably, the global self-attention network includes:
an input layer as an input to the global self-attention network;
the first convolution layer, the first maximum pooling layer and the transposition convolution layer are sequentially connected, and the input end of the first convolution layer is connected with the input layer;
the input end of the second convolution layer is connected with the input layer, and the output of the second maximum pooling layer and the output of the transposed convolution layer are subjected to feature fusion and then are used as the input of a classifier;
the input end of the third convolution layer is connected with the input layer, and the output of the third maximum pooling layer and the output of the classifier are subjected to feature fusion and then are used as the input of a fourth convolution layer;
and the input end of the fourth maximum pooling layer is connected with the input layer, and the output of the fourth maximum pooling layer and the output of the fourth convolutional layer are subjected to characteristic superposition to be used as the output of the global self-attention network.
Preferably, in step S22, arterial plaque identification is performed on each of the regions of interest by using a region candidate network, so as to obtain the plaque position information in each of the regions of interest as the image identification result.
Preferably, in step S3, a recurrent neural network is used to process each image identification result to obtain the arterial plaque identification result of the ultrasound scanning video.
Preferably, in step S1, after the ultrasound scanning video is acquired, image enhancement is performed on the ultrasound scanning video, and then each ultrasound scanning image is extracted from the ultrasound scanning video after image enhancement.
The invention also provides an arterial plaque identification system based on the ultrasonic scanning video, which applies the arterial plaque identification method and comprises the following steps:
the image extraction module is used for acquiring an ultrasonic scanning video and extracting continuous multi-frame ultrasonic scanning images from the ultrasonic scanning video;
the first identification module is connected with the image extraction module and used for respectively carrying out arterial plaque identification on each ultrasonic scanning image to obtain a corresponding image identification result;
and the second identification module is connected with the first identification module and used for processing each image identification result to obtain an artery plaque identification result of the ultrasonic scanning video.
Preferably, the first identification module includes:
the characteristic extraction unit is used for respectively extracting the characteristics of the ultrasonic scanning images to obtain interested areas in the ultrasonic scanning images;
and the position identification unit is connected with the feature extraction unit and is used for respectively carrying out arterial plaque identification on each region of interest to obtain plaque position information in each region of interest as the image identification result.
The technical scheme has the following advantages or beneficial effects: the artery plaque identification is carried out on the basis of the continuous multi-frame ultrasonic scanning images in the ultrasonic scanning video, the dynamic image information of the continuous frames can be captured, further, the artery plaque with the missing part and the artery plaque which is difficult to identify by a single ultrasonic scanning image can be well detected, the identification rate is improved, and the false positive detection rate is reduced.
Drawings
Fig. 1 is a schematic flow chart of an arterial plaque identification method based on an ultrasound scanning video according to a preferred embodiment of the present invention;
FIG. 2 is a schematic flow chart illustrating the process of performing arterial plaque identification on each ultrasound scanning image to obtain corresponding image identification results according to a preferred embodiment of the present invention;
fig. 3 is a schematic diagram of a main network structure adopted by the arterial plaque identification method in the preferred embodiment of the invention;
FIG. 4 is a schematic diagram of a network structure of a global self-attention network according to a preferred embodiment of the present invention;
fig. 5 is a schematic structural diagram of an arterial plaque identification system based on an ultrasound scanning video according to a preferred embodiment of the present invention.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments. The present invention is not limited to the embodiment, and other embodiments may be included in the scope of the present invention as long as the gist of the present invention is satisfied.
In accordance with the above problems in the prior art, there is provided an artery plaque identification method based on ultrasound scanning video, as shown in fig. 1, including:
step S1, acquiring an ultrasonic scanning video, and extracting continuous multiframe ultrasonic scanning images from the ultrasonic scanning video;
step S2, performing arterial plaque identification on each ultrasonic scanning image to obtain corresponding image identification results;
and step S3, processing each image recognition result to obtain an artery plaque recognition result of the ultrasonic scanning video.
Specifically, when ultrasound scanning is performed, missing arterial plaque in a single ultrasound scanning image obtained by scanning may be caused due to different scanning methods or scanning angles, or artery plaque in the single ultrasound scanning image obtained by scanning is difficult to identify and causes misdiagnosis.
In a preferred embodiment of the present invention, as shown in fig. 2, step S2 includes:
step S21, respectively extracting the characteristics of each ultrasonic scanning image to obtain an interested area in each ultrasonic scanning image;
step S22, performing arterial plaque identification on each region of interest to obtain plaque position information in each region of interest as an image identification result.
In a preferred embodiment of the present invention, before performing step S21, the method further includes dividing each ultrasound scanning image into a plurality of groups;
in step S21, feature extraction is performed on each group of ultrasound scanning images to obtain an area of interest corresponding to each group of ultrasound scanning images;
in step S22, arterial plaque identification is performed on the region of interest corresponding to each group of ultrasound scanning images, respectively, to obtain image identification results corresponding to each group of ultrasound scanning images.
Specifically, when the number of the extracted continuous ultrasound scanning images is large, the calculation speed may be slow due to a large calculation amount when the feature extraction is performed.
In a preferred embodiment of the present invention, in step S21, a global self-attention network is used to perform feature extraction on each group of ultrasound scanning images respectively to obtain an interested region in each group of ultrasound scanning images.
Specifically, in this embodiment, as shown in fig. 3, in the case of grouping consecutive ultrasound scanning images, each group of ultrasound scanning images is respectively configured with corresponding global self-attention network for parallel processing. For example, when extracting continuous 9 frames of ultrasound scanning images, the three groups of ultrasound scanning images can be divided into three groups, wherein the first group comprises continuous first three frames of ultrasound scanning images, the second group comprises continuous middle three frames of ultrasound scanning images, and the third group comprises continuous last three frames of ultrasound scanning images, wherein each group of ultrasound scanning images respectively performs feature extraction through a corresponding global self-attention network, and compared with the feature extraction of 9 frames of ultrasound scanning images through a global self-attention network, one global self-attention network only needs to perform feature extraction on 3 frames of ultrasound scanning images through grouping, so that the calculation amount is effectively reduced, and the operation speed is further improved. It will be appreciated that the number of the above-described extracted continuous ultrasound scan images and the number of the groups may be adjusted according to the requirements, and likewise, the number of the global self-attention network configurations may be adjusted according to the number of the groups.
In a preferred embodiment of the present invention, as shown in fig. 4, the global self-attention network comprises:
an input layer 100 as input to the global self-attention network;
the first convolution layer 200, the first maximum pooling layer 300 and the transposed convolution layer 400 are connected in sequence, wherein the input end of the first convolution layer 200 is connected with the input layer 100;
the second convolutional layer 201 and the second maximum pooling layer 301 are connected in sequence, the input end of the second convolutional layer 201 is connected with the input layer 100, and the output of the second maximum pooling layer 301 and the output of the transposed convolutional layer 400 are subjected to feature fusion and then serve as the input of a classifier 500;
the input end of the third convolutional layer 202 is connected with the input layer 100, and the output of the third max pooling layer 302 is subjected to feature fusion with the output of the classifier 500 and then is used as the input of a fourth convolutional layer 203;
and a fourth maximum pooling layer 303, wherein an input end of the fourth maximum pooling layer 303 is connected with the input layer 100, and an output of the fourth maximum pooling layer 303 and an output of the fourth convolutional layer 203 are subjected to characteristic superposition to be used as an output of the global self-attention network.
Specifically, in the present embodiment, the first convolution layer 200, the second convolution layer 201, and the fourth convolution layer 203 employ 1 × 1 convolution kernels, and the third convolution layer 202 employs 3 × 3 convolution kernels. The first largest pooling layer 300, the second largest pooling layer 301, the third largest pooling layer 302, and the fourth largest pooling layer 303 are all 2 x 2 largest pooling layers. The classifier 500 is a Softmax classifier. Furthermore, the effective information in each group of ultrasonic scanning images can be obtained through the global self-attention network, the pooling function is adopted for down-sampling, and 1 × 1 convolution layers are used for compressing channels, so that the inference efficiency can be effectively improved.
In a preferred embodiment of the present invention, in step S22, arterial plaque identification is performed on each region of interest by using a region candidate network to obtain plaque position information in each region of interest as an image identification result.
Specifically, in this embodiment, the Region candidate Network (RPN) may be capable of identifying plaque position information in an area of interest, as shown in fig. 3, the Region candidate Network is connected to the global self-attention Network, an output of the global self-attention Network is used as an input of the Region candidate Network, and when ultrasound scanning images of consecutive multiple frames are divided into three groups, three Region candidate networks may be correspondingly configured, and arterial plaque identification may be performed on the area of interest obtained by extracting features of each group of ultrasound scanning images, so as to effectively reduce a calculation amount, and further improve an operation speed.
In a preferred embodiment of the present invention, in step S3, a recurrent neural network is used to process each image recognition result to obtain an artery plaque recognition result of the ultrasound scanning video.
Specifically, in this embodiment, the Recurrent Neural Network (RNN) is a Neural Network capable of processing sequence data, and as shown in fig. 3, an input of the Recurrent Neural Network is an output of a candidate Network in each region, in order to obtain image recognition results of different groups and reduce information loss caused by using only a convolutional Network, the Recurrent Neural Network is used for performing remote information acquisition, and image recognition results corresponding to continuous multiple frames of ultrasound scanning images can be dynamically captured, so as to obtain an arterial plaque recognition result of an ultrasound scanning video.
In a preferred embodiment of the present invention, in step S1, after the acquiring the ultrasound scanning video, image enhancement is further performed on the ultrasound scanning video, and then each ultrasound scanning image is extracted from the ultrasound scanning video after the image enhancement.
Specifically, in the present embodiment, the manner of image enhancement includes, but is not limited to, operations such as overall rotation, scaling, and cutting of the entire video image.
The invention also provides an arterial plaque identification system based on the ultrasonic scanning video, which applies the arterial plaque identification method, and as shown in fig. 5, the arterial plaque identification system comprises:
the image extraction module 1 is used for acquiring an ultrasonic scanning video and extracting continuous multi-frame ultrasonic scanning images from the ultrasonic scanning video;
the first identification module 2 is connected with the image extraction module 1 and is used for respectively carrying out arterial plaque identification on each ultrasonic scanning image to obtain corresponding image identification results;
and the second identification module 3 is connected with the first identification module 2 and used for processing each image identification result to obtain an artery plaque identification result of the ultrasonic scanning video.
In a preferred embodiment of the present invention, the first identification module 2 comprises:
the feature extraction unit 21 is configured to perform feature extraction on each ultrasound scanning image to obtain an interested region in each ultrasound scanning image;
and the position identification unit 22 is connected with the feature extraction unit 21 and is used for respectively carrying out arterial plaque identification on each region of interest to obtain plaque position information in each region of interest as an image identification result.
While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention.
Claims (10)
1. An arterial plaque identification method based on an ultrasonic scanning video is characterized by comprising the following steps:
step S1, acquiring an ultrasonic scanning video, and extracting continuous multi-frame ultrasonic scanning images from the ultrasonic scanning video;
step S2, performing arterial plaque identification on each ultrasonic scanning image to obtain corresponding image identification results;
and step S3, processing each image recognition result to obtain an artery plaque recognition result of the ultrasonic scanning video.
2. The arterial plaque identification method according to claim 1, wherein said step S2 includes:
step S21, respectively extracting the features of each ultrasonic scanning image to obtain the region of interest in each ultrasonic scanning image;
step S22, performing arterial plaque identification on each region of interest to obtain plaque position information in each region of interest as the image identification result.
3. The arterial plaque identification method according to claim 2, wherein before the step S21, further comprising dividing each of the ultrasound scanning images into a plurality of groups;
in step S21, feature extraction is performed on each group of ultrasound scanning images to obtain the region of interest corresponding to each group of ultrasound scanning images;
in the step S22, arterial plaque identification is performed on the region of interest corresponding to each group of the ultrasound scanning images, respectively, to obtain the image identification result corresponding to each group of the ultrasound scanning images.
4. The method for identifying arterial plaque according to claim 2, wherein in the step S21, a global self-attention network is adopted to perform feature extraction on each group of the ultrasound scanning images respectively to obtain the region of interest in each group of the ultrasound scanning images.
5. The arterial plaque identification method of claim 4 wherein the global self-attention network comprises:
an input layer as an input to the global self-attention network;
the first convolution layer, the first maximum pooling layer and the transposition convolution layer are sequentially connected, and the input end of the first convolution layer is connected with the input layer;
the input end of the second convolution layer is connected with the input layer, and the output of the second maximum pooling layer and the output of the transposed convolution layer are subjected to feature fusion and then are used as the input of a classifier;
the input end of the third convolution layer is connected with the input layer, and the output of the third maximum pooling layer and the output of the classifier are subjected to feature fusion and then are used as the input of a fourth convolution layer;
and the input end of the fourth maximum pooling layer is connected with the input layer, and the output of the fourth maximum pooling layer and the output of the fourth convolutional layer are subjected to characteristic superposition to be used as the output of the global self-attention network.
6. The method according to claim 2, wherein in step S22, arterial plaque identification is performed on each region of interest using a region candidate network to obtain the plaque position information in each region of interest as the image identification result.
7. The method according to claim 1, wherein in step S3, a recurrent neural network is used to process each image recognition result to obtain the arterial plaque recognition result of the ultrasound scanning video.
8. The method for identifying arterial plaque according to claim 1, wherein in the step S1, after the obtaining of the ultrasound scanning video, the method further comprises performing image enhancement on the ultrasound scanning video, and then extracting each ultrasound scanning image from the image-enhanced ultrasound scanning video.
9. An arterial plaque identification system based on ultrasonic scanning video, which is characterized in that the arterial plaque identification method according to any one of claims 1-8 is applied, and the arterial plaque identification system comprises:
the image extraction module is used for acquiring an ultrasonic scanning video and extracting continuous multi-frame ultrasonic scanning images from the ultrasonic scanning video;
the first identification module is connected with the image extraction module and used for respectively carrying out arterial plaque identification on each ultrasonic scanning image to obtain a corresponding image identification result;
and the second identification module is connected with the first identification module and used for processing each image identification result to obtain an artery plaque identification result of the ultrasonic scanning video.
10. The arterial plaque identification system of claim 9 wherein the first identification module comprises:
the characteristic extraction unit is used for respectively extracting the characteristics of the ultrasonic scanning images to obtain interested areas in the ultrasonic scanning images;
and the position identification unit is connected with the feature extraction unit and is used for respectively carrying out arterial plaque identification on each region of interest to obtain plaque position information in each region of interest as the image identification result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111556641.1A CN114469174B (en) | 2021-12-17 | 2021-12-17 | Arterial plaque identification method and system based on ultrasonic scanning video |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111556641.1A CN114469174B (en) | 2021-12-17 | 2021-12-17 | Arterial plaque identification method and system based on ultrasonic scanning video |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114469174A true CN114469174A (en) | 2022-05-13 |
| CN114469174B CN114469174B (en) | 2024-10-18 |
Family
ID=81493266
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111556641.1A Active CN114469174B (en) | 2021-12-17 | 2021-12-17 | Arterial plaque identification method and system based on ultrasonic scanning video |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114469174B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116705252A (en) * | 2023-06-16 | 2023-09-05 | 脉得智能科技(无锡)有限公司 | Construction method, image classification method, device and medium for prostate cancer diagnosis model |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110257545A1 (en) * | 2010-04-20 | 2011-10-20 | Suri Jasjit S | Imaging based symptomatic classification and cardiovascular stroke risk score estimation |
| CN104392223A (en) * | 2014-12-05 | 2015-03-04 | 青岛科技大学 | Method for recognizing human postures in two-dimensional video images |
| 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 |
| CN110428417A (en) * | 2019-08-13 | 2019-11-08 | 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 |
| CN110946619A (en) * | 2019-11-27 | 2020-04-03 | 杨靖 | Intravascular ultrasonic automatic imaging omics analysis system and analysis method |
| WO2020087838A1 (en) * | 2018-10-29 | 2020-05-07 | 深圳先进技术研究院 | Blood vessel wall plaque recognition device, system and method, and storage medium |
| CN111950388A (en) * | 2020-07-22 | 2020-11-17 | 上海市同仁医院 | Vulnerable plaque tracking and identifying system and method |
| CN112016623A (en) * | 2020-08-28 | 2020-12-01 | 佳都新太科技股份有限公司 | Face clustering method, device, equipment and storage medium |
| CN112767252A (en) * | 2021-01-26 | 2021-05-07 | 电子科技大学 | Image super-resolution reconstruction method based on convolutional neural network |
| CN113192062A (en) * | 2021-05-25 | 2021-07-30 | 湖北工业大学 | Arterial plaque ultrasonic image self-supervision segmentation method based on image restoration |
-
2021
- 2021-12-17 CN CN202111556641.1A patent/CN114469174B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110257545A1 (en) * | 2010-04-20 | 2011-10-20 | Suri Jasjit S | Imaging based symptomatic classification and cardiovascular stroke risk score estimation |
| CN104392223A (en) * | 2014-12-05 | 2015-03-04 | 青岛科技大学 | Method for recognizing human postures in two-dimensional video images |
| WO2020087838A1 (en) * | 2018-10-29 | 2020-05-07 | 深圳先进技术研究院 | Blood vessel wall plaque recognition device, system and method, and storage medium |
| 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 |
| CN110428417A (en) * | 2019-08-13 | 2019-11-08 | 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 |
| CN110946619A (en) * | 2019-11-27 | 2020-04-03 | 杨靖 | Intravascular ultrasonic automatic imaging omics analysis system and analysis method |
| CN111950388A (en) * | 2020-07-22 | 2020-11-17 | 上海市同仁医院 | Vulnerable plaque tracking and identifying system and method |
| CN112016623A (en) * | 2020-08-28 | 2020-12-01 | 佳都新太科技股份有限公司 | Face clustering method, device, equipment and storage medium |
| CN112767252A (en) * | 2021-01-26 | 2021-05-07 | 电子科技大学 | Image super-resolution reconstruction method based on convolutional neural network |
| CN113192062A (en) * | 2021-05-25 | 2021-07-30 | 湖北工业大学 | Arterial plaque ultrasonic image self-supervision segmentation method based on image restoration |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116705252A (en) * | 2023-06-16 | 2023-09-05 | 脉得智能科技(无锡)有限公司 | Construction method, image classification method, device and medium for prostate cancer diagnosis model |
| CN116705252B (en) * | 2023-06-16 | 2024-05-31 | 脉得智能科技(无锡)有限公司 | Construction method, image classification method, device and medium for prostate cancer diagnosis model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114469174B (en) | 2024-10-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112258488A (en) | Medical image focus segmentation method | |
| CN111429474A (en) | Mammary gland DCE-MRI image focus segmentation model establishment and segmentation method based on mixed convolution | |
| CN116416497B (en) | Bearing fault diagnosis system and method | |
| CN103747270A (en) | Redundancy elimination method and system for capsule endoscope video image | |
| Liu et al. | Using the VQ-VAE to improve the recognition of abnormalities in short-duration 12-lead electrocardiogram records | |
| CN111915557A (en) | Deep learning atrial septal defect detection method and device | |
| Rajinikanth et al. | Hand-sketchs based Parkinson's disease screening using lightweight deep-learning with two-Fold training and fused optimal features | |
| Huang et al. | A deep segmentation network of stent structs based on IoT for interventional cardiovascular diagnosis | |
| CN114469174B (en) | Arterial plaque identification method and system based on ultrasonic scanning video | |
| CN114998303A (en) | Small intestine interstitial tumor detection method with strong feature extraction capability | |
| Diame et al. | Deep learning architiectures for aided melanoma skin disease recognition: a review | |
| CN110327034B (en) | Tachycardia electrocardiogram screening method based on depth feature fusion network | |
| CN111402207A (en) | Ultrasonic angiography video data analysis method based on composite neural network | |
| CN110811591A (en) | Heart failure grading method based on heart rate variability | |
| CN113689441A (en) | DeepLabV3 network-based left ventricle ultrasonic dynamic segmentation method | |
| Zang et al. | AmtNet: Attentional multi-scale temporal network for phonocardiogram signal classification | |
| CN116269426A (en) | Twelve-lead ECG-assisted heart disease multi-mode fusion screening method | |
| JP4982393B2 (en) | Image filtering apparatus, image filtering program, image filtering method, and ultrasonic diagnostic apparatus | |
| CN116452617A (en) | Whole heart segmentation method based on cross-modal domain transfer learning | |
| CN116058849A (en) | Atrial fibrillation detection method, device and equipment | |
| Hussein et al. | Pre-processing Importance for Extracting Contours from Noisy Echocardiographic Images | |
| Khudair et al. | Diabetes Diagnosis Using Deep Learning | |
| Bouhdiba et al. | Ultrasound Segmentation of Thyroid Nodules: An Enhancement Deep Neural Network Framework | |
| Rahman et al. | Attention Enabled MultiResUNet for Bio-Medical Image Segmentation | |
| Zhu et al. | DASENet: A Detail Aware U-Net with Shuffle Excitation for Retinal Vessel Segmentation |
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 |