CN112927212B - OCT cardiovascular plaque automatic identification and analysis method based on deep learning - Google Patents
OCT cardiovascular plaque automatic identification and analysis method based on deep learning Download PDFInfo
- Publication number
- CN112927212B CN112927212B CN202110263890.5A CN202110263890A CN112927212B CN 112927212 B CN112927212 B CN 112927212B CN 202110263890 A CN202110263890 A CN 202110263890A CN 112927212 B CN112927212 B CN 112927212B
- Authority
- CN
- China
- Prior art keywords
- image
- section
- result
- plaque
- cross
- 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
- 238000013135 deep learning Methods 0.000 title claims abstract description 20
- 238000004458 analytical method Methods 0.000 title claims abstract description 18
- 230000002526 effect on cardiovascular system Effects 0.000 title claims abstract description 17
- 230000011218 segmentation Effects 0.000 claims abstract description 52
- 238000012545 processing Methods 0.000 claims abstract description 26
- 238000000034 method Methods 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000003672 processing method Methods 0.000 claims description 5
- 238000013139 quantization Methods 0.000 abstract description 4
- 238000013461 design Methods 0.000 abstract description 2
- 238000012014 optical coherence tomography Methods 0.000 description 31
- 238000013527 convolutional neural network Methods 0.000 description 5
- 208000024172 Cardiovascular disease Diseases 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 208000029078 coronary artery disease Diseases 0.000 description 2
- 210000004351 coronary vessel Anatomy 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 150000002632 lipids Chemical class 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 208000037260 Atherosclerotic Plaque Diseases 0.000 description 1
- 101100508406 Caenorhabditis elegans ifa-1 gene Proteins 0.000 description 1
- 241001465754 Metazoa Species 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000004204 blood vessel Anatomy 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000005305 interferometry Methods 0.000 description 1
- 238000000386 microscopy Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000003287 optical effect 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- 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
- G06T2207/10101—Optical tomography; Optical coherence tomography [OCT]
-
- 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
-
- 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
Abstract
The invention provides an OCT cardiovascular plaque automatic identification and analysis method based on deep learning, which comprises the following steps: deep section treatment: dividing the depth section of the OCT image to obtain a division result A1; cross section treatment: dividing the cross section of the OCT image to obtain a division result A2; and (3) sectional comprehensive treatment: integrating the deep section processing result A1 and the cross section processing result A2 and performing segmentation processing to obtain a segmentation result A; analysis of results: and calculating plaque attenuation index IPA according to the segmentation result A to obtain plaque category. The invention has reasonable design, integrates different image classification characteristics and OCT clinical characteristics, and can effectively solve the problem of weak plaque identification capability in the prior art by carrying out three-dimensional reconstruction and segmentation on the 3D OCT image and obtaining the quantization index IPA and finally classifying the plaque based on the IPA.
Description
Technical Field
The invention relates to the field of image analysis, in particular to an OCT cardiovascular plaque automatic identification and analysis method based on deep learning.
Background
Cardiovascular diseases have developed as the first killer of resident disease death in China, causing immeasurable harm to national health and national economic development. Unstable lipid plaques are a major cause of the first-death cardiovascular disease, coronary heart disease.
Intravascular optical coherence tomography (IV-OCT) technology is rapidly developing into an effective means for diagnosing coronary heart disease due to its ultra-high imaging resolution (10-20 μm). IV-OCT images can clearly visualize different types of atherosclerotic plaques, known as "living tissue microscopy".
In segmenting plaques, N.Gessert et al pre-trains all models on ImageNet, then trains images using ResNet50-V2 and DenseNet121 models and polar and Cartesian coordinate systems to achieve segmentation and classification of several plaques for coronary OCT images, and attempts several different migration learning modes, finally comparing the serial results of several methods. Zhang et al propose a hybrid model of a faster-CNN network model with a fourth-order partial differential equation and a global local active contour model, first using the faster-CNN network model to detect calcified plaque areas, locating calcified plaque locations, setting a bounding box for a feature region, and setting the bounding box as an initial contour of the active contour model, and then minimizing the joint energy function of the fourth-order partial differential equation and the global local active contour model portion by a gradient descent and finite difference scheme, thereby segmenting calcified plaque in the coronary OCT image. He et al used OTSU automatic thresholding algorithm to segment tissue regions in the image, then input them as input sources into a CNN-based classifier, and divide focal tissue in the coronary OCT image into 5 different tissue categories, completing plaque characterization of the coronary OCT image. The problem of segmentation of coronary artery plaque is solved by using deep learning, a large data set is needed, or the data volume is increased through a large amount of data expansion, and the calculation amount and the requirement on hardware brought by a deep convolutional neural network are all difficulties of the deep learning in application to the segmentation of coronary artery OCT image plaque.
These methods are mainly classified into methods focusing on using deep learning, such as AlexNet, googleNet, VGG-Net, faster-CNN, denseNet, resNet, etc., in which plaque is segmented at each deep section, and then deep features are extracted from the segmented regions and then classified. These methods have achieved better results on image classification, such as ImageNet. They are mainly used for classifying and identifying different kinds of animals, plants, tools and other common articles of life. However, for OCT 3D images, the images produced by optical instrument interferometry, where plaque is typically small and 3D images, the ability to identify different plaques is typically weak, and there is a certain disadvantage.
Disclosure of Invention
Aiming at the defects of the prior art, the invention aims to provide an OCT cardiovascular plaque automatic identification and analysis method based on deep learning, which is used for carrying out three-dimensional reconstruction and segmentation on a 3D OCT image to obtain a quantization index IPA and finally classifying the plaque based on the IPA.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
an OCT cardiovascular plaque automatic identification and analysis method based on deep learning comprises the following steps:
s100, deep section processing: dividing the depth section of the OCT image to obtain a division result A1;
s200, cross section processing: dividing the cross section of the OCT image to obtain a division result A2;
s300, section comprehensive treatment: integrating the deep section processing result A1 and the cross section processing result A2 and performing segmentation processing to obtain a segmentation result A;
s400, analyzing results: and calculating plaque attenuation index IPA according to the segmentation result A to obtain plaque category.
In a preferred embodiment of the present invention, in step S100, the deep section processing specific steps include:
s101, inputting a 2D Image { Image (i, j, k) of a depth section; k=1, 2,3, …, M } = { Image (i, j, 1), image (i, j, 2), image (i, j, 3), …, image (i, j, M) }, wherein Image (i, j, k) represents coordinates of each point in the 2D Image;
s102, outputting the segmented 2D image, wherein the size of the segmented 2D image is consistent with that of the input image, the pixels of the useful area are 1, the pixels of the useless area are 0, and after obtaining the 2D images of all the depth sections, the 3D image A1 is obtained through combination.
In a preferred embodiment of the present invention, the specific processing method in step S102 is as follows: and carrying out convolution operation on each 2D image of the depth section, and combining to obtain a 3D image A1.
In a preferred embodiment of the present invention, in step S200, the cross-section processing specific steps include:
s201, inputting a 2D Image { Image (i, j, k) of a cross section; i=1, 2,3, …, N } = { Image (1, j, k), image (2, j, k), image (3, j, k), …, image (N, j, k) }, wherein Image (i, j, k) represents coordinates of each point in the 2D Image;
s202, outputting the segmented 2D image, wherein the size of the segmented 2D image is consistent with that of the input image, the pixels of the useful area are 1, the pixels of the useless area are 0, and after the 2D images of all cross sections are obtained, the 3D image A2 is obtained through combination.
In a preferred embodiment of the invention, the respective 2D images of the cross-section are convolved to obtain the 3D image A2 in combination.
In a preferred embodiment of the present invention, in step S300, the cross-section integration processing includes the specific steps of:
s301, inputting all depth section segmentation results A1 and cross section segmentation results A2;
s302, comprehensive depth interface divisionThe segmentation result A1 and the cross-section segmentation result A2 calculate a segmentation result A, wherein。
In a preferred embodiment of the present invention, in step S400, the specific steps of the result analysis include:
s401, calculating and inputting a segmentation result A according to a calculation formula of the segmentation result A;
s402, outputting plaque categories;
s403, calculating a value of plaque attenuation coefficient IPA:
，
where x is the plaque attenuation coefficient threshold, μ t For the attenuation coefficient, N (μ) t X) means the number of pixel points with attenuation coefficient larger than x, N total The effective area of the segmentation result a is shown.
Through the technical scheme, the invention has the following beneficial effects:
the invention has reasonable design, integrates different image classification characteristics and OCT clinical characteristics, and can effectively solve the problem of weak plaque identification capability in the prior art by carrying out three-dimensional reconstruction and segmentation on the 3D OCT image and obtaining the quantization index IPA and finally classifying the plaque based on the IPA.
Drawings
FIG. 1 is a flow chart of an OCT cardiovascular plaque automatic identification and analysis method based on deep learning;
FIG. 2 is a graph showing the calculation process of the segmentation result A1 of the OCT cardiovascular plaque automatic recognition and analysis method based on deep learning;
FIG. 3 is a graph showing the calculation process of the segmentation result A2 of the OCT cardiovascular plaque automatic recognition and analysis method based on deep learning;
FIG. 4 is a deep cross-sectional view of an OCT image of an OCT cardiovascular plaque automatic identification and analysis method based on deep learning according to the present invention;
fig. 5 is a cross-sectional view of an OCT image of a method for automatically identifying and analyzing cardiovascular plaques based on deep learning according to the present invention.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. As used in this disclosure and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
Referring to fig. 1, an embodiment of an OCT cardiovascular plaque automatic identification and analysis method based on deep learning according to the present invention is shown. In this embodiment, the plaque automatic identification and analysis method includes the following steps:
referring to fig. 1 and 2, step S100 is described in detail:
s100, deep section processing: and (3) performing segmentation processing on the depth section of the OCT image to obtain a segmentation result A1. The processing method of the segmentation result A1 comprises the following steps:
i. inputting a 2D Image { Image (i, j, k) of a depth section; k=1, 2,3, …, M } = { Image (i, j, 1), image (i, j, 2), image (i, j, 3), …, image (i, j, M) }, wherein Image (i, j, k) represents coordinates of each point in the 2D Image;
outputting the segmented 2D image, wherein the size of the segmented 2D image is consistent with that of the input image, the pixels of the useful area are 1, the pixels of the useless area are 0, and the 3D image A1 is obtained by combining the 2D images of all the depth sections. And performing convolution operation on each 2D image of the depth section, and combining to obtain a 3D image A1.
The specific treatment method is as follows:
Block1：
convolving a 2D Image (i, j, k) of a kth depth section, wherein k=1, 2,3, …, M, with a filter_k1 of 3*3 to obtain a result_k1, convolving a filter_k2 of 3*5 to obtain a result_k2, convolving a filter_k3 of 5*3 to obtain a result_k3, convolving a filter_k4 of 5*5 to obtain a result_k4, a1×result_k1+a2×result_k2+a3×result_k3+a4×result_k4=result_k1.
Block_k_2, block_k_3, block_k_4: the structure is the same as block_k_1.
Result k B1 passes through blockk2=result k B2,
result k B2 passes through blockk3=result k B3,
result k B3 passes through blockk4=result k B4,
c1*Result_k_B1+c2*Result_k_B2+c3*Result_k_B3+c4*Result_k_B4=A1。
referring to fig. 1 and 3, step S200 is described in detail:
s200, cross section processing: and performing segmentation processing on the cross section of the OCT image to obtain a segmentation result A2. The processing method of the segmentation result A2 comprises the following steps:
i. inputting a 2D Image { Image (i, j, k) of the cross section; i=1, 2,3, …, N } = { Image (1, j, k), image (2, j, k), image (3, j, k), …, image (N, j, k) }, wherein Image (i, j, k) represents coordinates of each point in the 2D Image;
outputting the segmented 2D image, wherein the size of the segmented 2D image is consistent with that of the input image, the pixels of the useful area are 1, the pixels of the useless area are 0, and after the 2D images of all cross sections are obtained, the 3D image A2 is obtained through combination. Wherein, convolution operation is carried out on each 2D image of the cross section, and the 3D images A2 are obtained in a combined way.
The specific method is as follows:
Block1:
convolving the i-th cross-section 2D Image { Image (i, j, k), i=1, 2,3, …, N }, performing a convolution operation on the filter_i1 of 3*3 to obtain result_i1, performing a convolution operation on the filter_i2 of 3*5 to obtain result_i2, performing a convolution operation on the filter_i3 of 5*3 to obtain result_i3, performing a convolution operation on the filter_i4 of 5*5 to obtain result_i4, b1×result_i1+b2×result_i2+b3×result_i3+b4×result_i4=result_i_b1.
Block2, block3, block4: the structure is the same as Block 1.
Result i B1 passes through block2=result i B2,
result i B2 passes through block3=result i B3,
result i B3 passes through block4=result i B4,
d1*Result_i_B1+d2*Result_i_B2+d3*Result_i_B3+d4*Result_i_B4=A2。
referring to fig. 1, step S300 is described in detail:
s300, section comprehensive treatment: and integrating the depth section processing result A1 and the cross section processing result A2, and performing segmentation processing to obtain a segmentation result A. The processing method of the segmentation result A comprises the following steps:
i. inputting all the depth section segmentation results A1 and the cross section segmentation results A2;
ii, integrating the depth interface segmentation result A1 and the cross section segmentation result A2 to calculate a segmentation result A, wherein。
The specific algorithm is as follows:
A=zeros(width=A1.width，height=A1.height，length=A1.length)
forifrom0toA1.width
forjfrom0toA1.height
forkfrom0toA1.length
ifA1(i，j，k)belongstoA2:
A(i，j，k)=A1(i，j，k)。
referring to fig. 1, step S400 is described in detail:
s400, analyzing results: and calculating plaque attenuation index IPA according to the segmentation result A to obtain plaque category. The specific method for analyzing the result is as follows:
i. calculating and inputting a segmentation result A according to a calculation formula of the segmentation result A;
outputting plaque categories;
calculating the value of plaque attenuation coefficient IPA:
，
where x is the plaque attenuation coefficient threshold, μ t For the attenuation coefficient, N (μ) t X) means the number of pixel points with attenuation coefficient larger than x, N total The effective area of the segmentation result a is shown. IPA represents plaque attenuation index (index) to quantify the degree of plaque attenuation in blood vessels. The calculation method is that when the pixel fraction of different types of plaques in the attenuation map is larger than the attenuation coefficient threshold value x, the pixel fraction is multiplied by 1000. The present study represents fibrous plaque with calculated value IPA6, IPA11 indicating lipid plaque. Among them, the larger the IPA, the worse the plaque stability.
The invention considers the characteristics of common image classification, combines the clinical characteristics of OCT, and simultaneously analyzes two sections of OCT images: deep section and cross section. Fig. 4 and 5 are 2D images of the depth cross section and the cross section of OCT, respectively. On the depth section: each 2D cross-section of the OCT image presents a circular or quasi-circular interference aperture; on the cross section, the OCT image can be viewed from the cross section after three-dimensional reconstruction, and each cross section is a 2D image and is specifically characterized in that a certain area in the middle is provided with a transverse bent interval stripe. The problem of weak plaque identification capability in the prior art can be effectively solved by performing three-dimensional reconstruction and segmentation on the 3D OCT image and obtaining a quantization index IPA and finally classifying the plaque based on the IPA.
While the invention has been described with respect to preferred embodiments thereof, it will be understood by those skilled in the art that various modifications and additions may be made without departing from the scope of the invention. Equivalent embodiments of the present invention will be apparent to those skilled in the art having the benefit of the teachings disclosed herein, when considered in the light of the foregoing disclosure, and without departing from the spirit and scope of the invention; meanwhile, any equivalent changes, modifications and evolution of the above embodiments according to the essential technology of the present invention still fall within the scope of the technical solution of the present invention.
Claims (5)
1. The OCT cardiovascular plaque automatic identification and analysis method based on deep learning is characterized by comprising the following steps of:
s100, deep section processing: dividing the depth section of the OCT image, and carrying out four convolution operations on the 2D image of the depth section to obtain a division result A1;
s200, cross section processing: dividing the cross section of the OCT image, and performing convolution operation on the 2D image of the cross section for four times to obtain a division result A2;
s300, section comprehensive treatment: integrating the deep section processing result A1 and the cross section processing result A2 and performing segmentation processing to obtain a segmentation result A;
in step S300, the section integrated processing specifically includes:
s301, inputting all depth section segmentation results A1 and cross section segmentation results A2;
s302, calculating a segmentation result A by integrating the depth interface segmentation result A1 and the cross section segmentation result A2, wherein；
S400, analyzing results: calculating plaque attenuation index IPA according to the segmentation result A to obtain plaque category;
in step S400, the specific steps of the result analysis include:
s401, calculating and inputting a segmentation result A according to a calculation formula of the segmentation result A;
s402, outputting plaque categories;
s403, calculating a value of plaque attenuation coefficient IPA:
，
where x is the plaque attenuation coefficient threshold, μ t For the attenuation coefficient, N (μ) t X) means the number of pixel points with attenuation coefficient larger than x, N total The effective area of the segmentation result a is shown.
2. The method for automatically identifying and analyzing cardiovascular plaques based on deep learning OCT according to claim 1, wherein in step S100, the deep cross-section processing comprises the specific steps of:
s101, inputting a 2D Image { Image (i, j, k) of a depth section; k=1, 2,3, …, M } = { Image (i, j, 1), image (i, j, 2), image (i, j, 3), …, image (i, j, M) }, wherein Image (i, j, k) represents coordinates of each point in the 2D Image;
s102, outputting the segmented 2D image, wherein the size of the segmented 2D image is consistent with that of the input image, the pixels of the useful area are 1, the pixels of the useless area are 0, and after obtaining the 2D images of all the depth sections, the 3D image A1 is obtained through combination.
3. The method for automatically identifying and analyzing OCT cardiovascular plaque based on deep learning according to claim 2, wherein the specific processing method in step S102 is as follows: and carrying out convolution operation on each 2D image of the depth section, and combining to obtain a 3D image A1.
4. The method for automatic recognition and analysis of deep learning based OCT cardiovascular plaque of claim 1, wherein in step S200, the cross-section processing comprises the specific steps of:
s201, inputting a 2D Image { Image (i, j, k) of a cross section; i=1, 2,3, …, N } = { Image (1, j, k), image (2, j, k), image (3, j, k), …, image (N, j, k) }, wherein Image (i, j, k) represents coordinates of each point in the 2D Image;
s202, outputting the segmented 2D image, wherein the size of the segmented 2D image is consistent with that of the input image, the pixels of the useful area are 1, the pixels of the useless area are 0, and after the 2D images of all cross sections are obtained, the 3D image A2 is obtained through combination.
5. The automatic recognition and analysis method of cardiovascular plaques based on deep learning of claim 1, wherein the respective 2D images of the cross-sections are convolved to obtain a 3D image A2 in combination.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110263890.5A CN112927212B (en) | 2021-03-11 | 2021-03-11 | OCT cardiovascular plaque automatic identification and analysis method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110263890.5A CN112927212B (en) | 2021-03-11 | 2021-03-11 | OCT cardiovascular plaque automatic identification and analysis method based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112927212A CN112927212A (en) | 2021-06-08 |
| CN112927212B true CN112927212B (en) | 2023-10-27 |
Family
ID=76172610
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110263890.5A Active CN112927212B (en) | 2021-03-11 | 2021-03-11 | OCT cardiovascular plaque automatic identification and analysis method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112927212B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113538471B (en) * | 2021-06-30 | 2023-09-22 | 上海联影医疗科技股份有限公司 | Plaque segmentation method, plaque segmentation device, computer equipment and storage medium |
| CN114469337A (en) * | 2021-07-05 | 2022-05-13 | 深圳市中科微光医疗器械技术有限公司 | Ablation catheter assembly, laser ablation system and method |
| CN113538365A (en) * | 2021-07-13 | 2021-10-22 | 深圳市中科微光医疗器械技术有限公司 | Method and device for calculating IPA of intra-cavity OCT image |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101011260A (en) * | 2006-02-01 | 2007-08-08 | 西门子公司 | Method and CT system for detecting and differentiating plaque in vessel structures of a patient |
| CN104398271A (en) * | 2014-11-14 | 2015-03-11 | 西安交通大学 | Method using three-dimensional mechanics and tissue specific imaging of blood vessels and plaques for detection |
| CN104794708A (en) * | 2015-04-10 | 2015-07-22 | 浙江工业大学 | Atherosclerosis plaque composition dividing method based on multi-feature learning |
| CN109697459A (en) * | 2018-12-04 | 2019-04-30 | 云南大学 | One kind is towards optical coherence tomography image patch Morphology observation method |
| CN110136157A (en) * | 2019-04-09 | 2019-08-16 | 华中科技大学 | A kind of three-dimensional carotid ultrasound image vascular wall dividing method based on deep learning |
| CN110223781A (en) * | 2019-06-03 | 2019-09-10 | 中国医科大学附属第一医院 | A kind of various dimensions plaque rupture Warning System |
| CN111161216A (en) * | 2019-12-09 | 2020-05-15 | 杭州脉流科技有限公司 | Intravascular ultrasound image processing method, device, equipment and storage medium based on deep learning |
| CN111445473A (en) * | 2020-03-31 | 2020-07-24 | 复旦大学 | Vascular membrane accurate segmentation method and system based on intravascular ultrasound image sequence multi-angle reconstruction |
| CN111768403A (en) * | 2020-07-09 | 2020-10-13 | 成都全景恒升科技有限公司 | Calcified plaque detection decision-making system and device based on artificial intelligence algorithm |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3363350B1 (en) * | 2009-09-23 | 2019-12-11 | Lightlab Imaging, Inc. | Lumen morphology and vascular resistance measurements data collection systems, apparatus and methods |
| US9693755B2 (en) * | 2012-01-19 | 2017-07-04 | Konica Minolta, Inc. | Ultrasound diagnostic device and method for controlling ultrasound diagnostic device |
| US20140303499A1 (en) * | 2013-04-09 | 2014-10-09 | Konica Minolta, Inc. | Ultrasound diagnostic apparatus and method for controlling the same |
| US11432719B2 (en) * | 2019-03-14 | 2022-09-06 | Oregon Health & Science University | Visual field simulation using optical coherence tomography and optical coherence tomographic angiography |
-
2021
- 2021-03-11 CN CN202110263890.5A patent/CN112927212B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101011260A (en) * | 2006-02-01 | 2007-08-08 | 西门子公司 | Method and CT system for detecting and differentiating plaque in vessel structures of a patient |
| CN104398271A (en) * | 2014-11-14 | 2015-03-11 | 西安交通大学 | Method using three-dimensional mechanics and tissue specific imaging of blood vessels and plaques for detection |
| CN104794708A (en) * | 2015-04-10 | 2015-07-22 | 浙江工业大学 | Atherosclerosis plaque composition dividing method based on multi-feature learning |
| CN109697459A (en) * | 2018-12-04 | 2019-04-30 | 云南大学 | One kind is towards optical coherence tomography image patch Morphology observation method |
| CN110136157A (en) * | 2019-04-09 | 2019-08-16 | 华中科技大学 | A kind of three-dimensional carotid ultrasound image vascular wall dividing method based on deep learning |
| CN110223781A (en) * | 2019-06-03 | 2019-09-10 | 中国医科大学附属第一医院 | A kind of various dimensions plaque rupture Warning System |
| CN111161216A (en) * | 2019-12-09 | 2020-05-15 | 杭州脉流科技有限公司 | Intravascular ultrasound image processing method, device, equipment and storage medium based on deep learning |
| CN111445473A (en) * | 2020-03-31 | 2020-07-24 | 复旦大学 | Vascular membrane accurate segmentation method and system based on intravascular ultrasound image sequence multi-angle reconstruction |
| CN111768403A (en) * | 2020-07-09 | 2020-10-13 | 成都全景恒升科技有限公司 | Calcified plaque detection decision-making system and device based on artificial intelligence algorithm |
Non-Patent Citations (5)
| Title |
|---|
| Molecular pathobiology of scleritis and its therapeutic implications;Undurti N Das;;International Journal of Ophthalmology(第01期);全文 * |
| OCT纵向图像形成分析;曾绍群，骆清铭，刘贤德，徐海峰，李再光;电子学报(第10期);全文 * |
| Optical coherence tomography angiography of the peripapillary region and macula in normal, primary open angle glaucoma, pseudoexfoliation glaucoma and ocular hypertension eyes;Helin Ceren Kse;Oya Tekeli;;International Journal of Ophthalmology(第05期);全文 * |
| 光学相干层析成像技术原理及研究进展;陆冬筱;房文汇;李玉瑶;李金华;王笑军;;中国光学(第05期);全文 * |
| 基于深度学习的颈动脉超声图像斑块分割算法;蔡梦媛;周然;程新耀;丁明跃;;生命科学仪器(第03期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112927212A (en) | 2021-06-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112927212B (en) | OCT cardiovascular plaque automatic identification and analysis method based on deep learning | |
| CN109166124B (en) | Retinal blood vessel morphology quantification method based on connected region | |
| CN108961261B (en) | Optic disk region OCT image hierarchy segmentation method based on space continuity constraint | |
| EP2285266B1 (en) | Automatic cup-to-disc ratio measurement system | |
| US7992999B2 (en) | Automated assessment of optic nerve head with spectral domain optical coherence tomography | |
| US8111896B2 (en) | Method and system for automatic recognition of preneoplastic anomalies in anatomic structures based on an improved region-growing segmentation, and commputer program therefor | |
| TWI544898B (en) | Device and method for determining a skin inflammation score | |
| CN113420826B (en) | Liver focus image processing system and image processing method | |
| JP2005199040A (en) | Method and apparatus for processing image data to aid in detecting disease | |
| Chen et al. | Automated ventricular systems segmentation in brain CT images by combining low-level segmentation and high-level template matching | |
| CN113610808B (en) | Group brain map individuation method, system and equipment based on individual brain connection diagram | |
| US20150371400A1 (en) | Segmentation and identification of closed-contour features in images using graph theory and quasi-polar transform | |
| CN107895364B (en) | A kind of three-dimensional reconstruction system for the preoperative planning of virtual operation | |
| JP2009531117A (en) | Method and system for automatically recognizing precancerous abnormalities in anatomical structures and corresponding computer program | |
| CN111415324B (en) | Classification and identification method for brain disease focus image space distribution characteristics based on magnetic resonance imaging | |
| US20070049839A1 (en) | System and method for automated airway evaluation for multi-slice computed tomography (msct) image data using airway lumen diameter, airway wall thickness and broncho-arterial ratio | |
| CN108416793B (en) | Choroidal vessel segmentation method and system based on three-dimensional coherence tomography image | |
| Dodo et al. | Graph-Cut Segmentation of Retinal Layers from OCT Images. | |
| Bodzioch et al. | New approach to gallbladder ultrasonic images analysis and lesions recognition | |
| Xu et al. | Weakly supervised detection of central serous chorioretinopathy based on local binary patterns and discrete wavelet transform | |
| CN112638262A (en) | Similarity determination device, method, and program | |
| CN116725563B (en) | Eyeball salience measuring device | |
| KR101514795B1 (en) | Quantification method of vessel | |
| Niessen et al. | Error metrics for quantitative evaluation of medical image segmentation | |
| CN115969400A (en) | Apparatus for measuring area of eyeball protrusion |
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 |