CN113707312A - Blood vessel quantitative identification method and device based on deep learning - Google Patents
Blood vessel quantitative identification method and device based on deep learning Download PDFInfo
- Publication number
- CN113707312A CN113707312A CN202111089382.6A CN202111089382A CN113707312A CN 113707312 A CN113707312 A CN 113707312A CN 202111089382 A CN202111089382 A CN 202111089382A CN 113707312 A CN113707312 A CN 113707312A
- Authority
- CN
- China
- Prior art keywords
- blood vessel
- network
- image data
- data
- perfusion
- 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.)
- Pending
Links
- 210000004204 blood vessel Anatomy 0.000 title claims abstract description 61
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000013135 deep learning Methods 0.000 title claims abstract description 19
- 230000010412 perfusion Effects 0.000 claims abstract description 54
- 238000012549 training Methods 0.000 claims abstract description 35
- 238000013526 transfer learning Methods 0.000 claims abstract description 31
- 238000013528 artificial neural network Methods 0.000 claims abstract description 16
- 238000002372 labelling Methods 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 7
- 238000005457 optimization Methods 0.000 claims description 2
- 238000003672 processing method Methods 0.000 abstract description 2
- 230000002107 myocardial effect Effects 0.000 description 23
- 238000012545 processing Methods 0.000 description 4
- 238000011158 quantitative evaluation Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 238000013508 migration Methods 0.000 description 3
- 230000005012 migration Effects 0.000 description 3
- 210000003484 anatomy Anatomy 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 238000002059 diagnostic imaging Methods 0.000 description 2
- 230000002526 effect on cardiovascular system Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000008014 freezing Effects 0.000 description 2
- 238000007710 freezing Methods 0.000 description 2
- 208000010125 myocardial infarction Diseases 0.000 description 2
- 208000006011 Stroke Diseases 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000002547 anomalous effect Effects 0.000 description 1
- 206010008118 cerebral infarction Diseases 0.000 description 1
- 230000003788 cerebral perfusion Effects 0.000 description 1
- 208000026106 cerebrovascular disease Diseases 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000008451 emotion Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- 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
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H30/00—ICT specially adapted for the handling or processing of medical images
- G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10088—Magnetic resonance imaging [MRI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10104—Positron emission tomography [PET]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30101—Blood vessel; Artery; Vein; Vascular
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Physics & Mathematics (AREA)
- Public Health (AREA)
- Biomedical Technology (AREA)
- Theoretical Computer Science (AREA)
- Primary Health Care (AREA)
- General Physics & Mathematics (AREA)
- Epidemiology (AREA)
- Radiology & Medical Imaging (AREA)
- Data Mining & Analysis (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Pathology (AREA)
- Computational Linguistics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biophysics (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
A method and a device for blood vessel quantitative identification based on deep learning are provided, wherein the method comprises the following steps: constructing a transfer learning network, inputting natural image data into the transfer learning network, and performing network pre-training to obtain a pre-trained recognition network; carrying out transfer learning training on the pre-trained recognition network by using multi-modal data to obtain a blood vessel quantitative recognition network; optimizing the blood vessel quantitative recognition network by using perfusion CT image data to obtain a blood vessel quantitative recognition neural network; inputting the collected target perfusion CT image into a blood vessel quantitative recognition neural network to obtain a blood vessel quantitative recognition result; a user inputs the collected perfusion CT image into the blood vessel quantitative recognition neural network, so that a blood vessel quantitative recognition result can be obtained, and compared with the traditional processing method, the method is higher in efficiency and better in robustness.
Description
Technical Field
The invention relates to the technical field of medical image processing, in particular to a method and a device for quantitatively identifying blood vessels based on deep learning.
Background
The clinical routine CT diagnosis usually generates a large amount of CT image data, so the image reading workload of imaging doctors is very large, while the manual reading usually accompanies subjective factors, and different results can be obtained because the emotions and states of the doctors after long-time reading or different clinical experiences.
Therefore, how to provide a method for automatically identifying a lesion area in an image to improve the robustness and speed of a medical image processing result is a problem to be solved by those skilled in the art.
Disclosure of Invention
The embodiment of the application provides a method and a device for quantitatively identifying blood vessels based on deep learning, and aims to solve the problems of poor robustness and low efficiency of medical image processing in the prior art.
In a first aspect, the present application provides a method for vessel quantitative identification based on deep learning, the method comprising:
constructing a transfer learning network, inputting natural image data into the transfer learning network, and performing network pre-training to obtain a pre-trained recognition network;
carrying out transfer learning training on the pre-trained recognition network by using multi-modal data to obtain a blood vessel quantitative recognition network;
optimizing the blood vessel quantitative recognition network by using perfusion CT image data to obtain a blood vessel quantitative recognition neural network;
and inputting the collected target perfusion CT image into a blood vessel quantitative recognition neural network to obtain a blood vessel quantitative recognition result.
In one embodiment, the pair of natural image data includes natural image data and scene recognition results.
In one embodiment, the multi-modality data pair includes multi-modality image data and vessel region identification data.
In one embodiment, the vessel region identification data is: and inputting the multi-modal image data after the vessel region is labeled by an expert or a doctor into the pre-trained recognition network for training to obtain the data.
In one embodiment, perfusion of a CT image data pair includes: perfusion CT image data and vessel region labeling data.
In one embodiment, the vessel region labeling data is: and inputting perfusion CT image data after vessel region labeling by a specialist or doctor into the pre-trained recognition network for training to obtain data.
In a second aspect, the present application further provides an apparatus for vessel quantitative identification based on deep learning, the apparatus comprising:
the pre-training unit is used for constructing a multi-mode transfer learning network, inputting the natural image data pair into the transfer learning network, and performing network pre-training to obtain a pre-trained recognition network;
the transfer learning training unit is used for carrying out transfer learning training on the pre-trained recognition network through multi-mode data to obtain a blood vessel quantitative recognition network;
the optimization unit is used for optimizing the blood vessel quantitative recognition network through perfusion CT image data to obtain a blood vessel quantitative recognition neural network;
and the result output unit is used for inputting the acquired target perfusion CT image into the blood vessel quantitative recognition neural network to obtain a blood vessel quantitative recognition result.
In a third aspect, the present application further provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for blood vessel quantitative identification based on deep learning according to the first aspect.
In a fourth aspect, the present application further provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the processor is caused to execute the method for vessel quantitative identification based on deep learning of the first aspect.
According to the method for quantitatively identifying the blood vessel based on deep learning, a transfer learning network is constructed, natural image data are input into the transfer learning network, network pre-training is carried out, and a pre-trained identification network is obtained; carrying out transfer learning training on the pre-trained recognition network by using multi-modal data to obtain a blood vessel quantitative recognition network; optimizing the blood vessel quantitative recognition network by using perfusion CT image data to obtain a blood vessel quantitative recognition neural network; inputting the collected target perfusion CT image into a blood vessel quantitative recognition neural network to obtain a blood vessel quantitative recognition result; a user inputs the collected perfusion CT image into the blood vessel quantitative recognition neural network, so that a blood vessel quantitative recognition result can be obtained, and compared with the traditional processing method, the method is higher in efficiency and better in robustness.
Drawings
For better clarity of the technical solutions of the embodiments of the present application, the drawings needed to be used in the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a flowchart of a method for blood vessel quantitative identification based on deep learning according to an embodiment of the present application;
fig. 2 is a flowchart of a method for quantitative identification of CT myocardial vessels according to an embodiment of the present application.
Detailed Description
The following further describes embodiments of the present invention with reference to the drawings. It should be noted that the description of the embodiments is provided to help understanding of the present invention, but the present invention is not limited thereto. In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
Referring to fig. 1, a flowchart of a method for blood vessel quantitative identification based on deep learning is shown, which includes:
s101, constructing a transfer learning network, inputting natural image data into the transfer learning network, and performing network pre-training to obtain a pre-trained recognition network.
In an embodiment, the pair of natural image data includes natural image data and scene recognition results.
S102, carrying out transfer learning training on the recognition network after pre-training by using the multi-modal data to obtain the blood vessel quantitative recognition network.
In an embodiment, the multi-modality data pair includes multi-modality image data and vessel region identification data.
In one embodiment, the vessel region identification data is: and inputting the multi-modal image data after the vessel region is labeled by an expert or a doctor into the pre-trained recognition network for training to obtain the data.
S103, the perfusion CT image data is used for optimizing the blood vessel quantitative recognition network to obtain a blood vessel quantitative recognition neural network.
In one embodiment, the perfusion CT image data pair includes: perfusion CT image data and vessel region labeling data.
In one embodiment, the vessel region labeling data is: and inputting perfusion CT image data after vessel region labeling by a specialist or doctor into the pre-trained recognition network for training to obtain data.
And S104, inputting the acquired target perfusion CT image into a blood vessel quantitative recognition neural network to obtain a blood vessel quantitative recognition result.
In one embodiment, the natural image data pair is a "natural image data-image tag" dataset comprising natural image data and scene recognition results;
the multi-modality data pair labels the "medical image data with anatomical relevance-vessel region" dataset, which may be an MRI image and/or a PET image of the same scanning region as the CT perfusion image; the multi-modal image data comprises PET cardiovascular and cerebrovascular perfusion image data and MRI cardiovascular and cerebrovascular perfusion image data.
In one embodiment, the network pre-training of the natural image data pair input recognition network comprises:
and (3) reconstructing perfusion CT projection data by using perfusion CT data, then performing migration learning on the blood vessel quantitative recognition network, and training to obtain the recognition network capable of performing quantitative perfusion CT image cerebral infarction region.
In an embodiment, before performing the transfer learning training, the method further includes making a transfer learning strategy, specifically:
according to the size and the number of parameters of the perfusion CT data set, two learning strategies can be selected for transfer learning: (1) freezing (FROZEN) the parameters of the first n layers, namely, not changing the values of the n layers when training the quantitative evaluation network of the perfusion CT blood vessel region; (2) the first n layer parameters are not frozen but the values of the network parameters are continuously adjusted, called fine-tune (fine-tune).
In the transfer learning strategy making, the training network realizes the fusion of the correlation of anatomical structures and the diversity of image information of different modes in PET, MRI and CT images of a scanning region, and optimizes the quantitative evaluation network parameters of a blood vessel region.
In one embodiment, the construction method of the blood vessel quantitative identification network specifically comprises the following steps:
first, a training set is defined as S { (X)i,Yi) 1, 2, 3, aiFor the ith input image, YiE {0, 1} is the label for the ith input image, where Yi1 is defined as an abnormal image, Yi0 is defined as a non-anomalous image. Simultaneous definition ofIs the probability of the kth pixel of the ith input image, where k ═ 1, 2, 3i|},|Xi| represents XiTotal number of pixels.Probability map (probability) for image level predictionmap) that can pass all of the pixel levelsAnd calculating to obtain the cost function:
Different from the traditional information processing system, the deep learning technology can be used for directly inputting data for training and then making prediction by constructing a network model, so that the manual feature extraction and the image processing technology design can be avoided.
Different medical imaging modalities (such as CT, MRI, PET and the like) can provide different medical image information, and at the same time, the medical imaging modalities share similar anatomical structure characteristic information, so that the multi-modal characteristic information can be internally shared, the diagnosis effect is improved, therefore, the scheme can also be used for the quantitative identification of CT myocardial vessels,
referring to fig. 2, a flowchart of a method for quantitative identification of CT myocardial vessels is shown in an embodiment, which includes:
s201, inputting natural image data to construct a multi-mode migration learning network.
In an embodiment, the pair of natural image data includes natural image data and scene recognition results.
S202, carrying out transfer learning training on the transfer learning network after the natural image data pair training by the myocardial perfusion PET data pair to obtain a blood vessel region quantitative evaluation and identification network.
In an embodiment, the myocardial perfusion PET data pair comprises myocardial perfusion PET image data and a myocardial perfusion PET image vessel region annotation.
S203, a transfer learning strategy is formulated, myocardial perfusion CT projection data are firstly reconstructed by myocardial perfusion CT data, then transfer learning is carried out on the blood vessel quantitative identification network, and blood vessel region quantitative identification network parameters are optimized.
In an embodiment, the myocardial perfusion CT data pair comprises myocardial perfusion CT projection data and myocardial perfusion CT image vessel region labels.
In one embodiment, the vessel region labeling of the myocardial perfusion PET image and the vessel region labeling of the myocardial perfusion CT image are the myocardial perfusion PET image data and the myocardial perfusion CT image data after the vessel region labeling of the expert or the doctor.
The myocardial perfusion CT data pair, namely the size and the parameter number of the myocardial perfusion CT projection data and the myocardial perfusion CT image vessel region label, can be selected as follows: (1) freezing (FROZEN) the parameters of the first n layers, namely not changing the values of the n layers when training the myocardial perfusion CT blood vessel region quantitative evaluation identification network (203); (2) the first n layers of parameters are not frozen, and the values of the network parameters are continuously adjusted, so that two strategies called fine-tune (fine-tune) are used for migration learning. Finally, we obtain an identification network (104) that can enable quantitative myocardial perfusion CT images of stroke regions.
S204, collecting the myocardial perfusion CT projection data to be identified, inputting the collected perfusion CT projection data to be identified into a recognition network of the trained quantitative cerebral perfusion CT image myocardial infarction region, and outputting the focus recognition result as the myocardial perfusion CT image myocardial infarction region.
The embodiments of the present invention have been described in detail with reference to the accompanying drawings, but the present invention is not limited to the described embodiments. It will be apparent to those skilled in the art that various changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, and the scope of protection is still within the scope of the invention.
Claims (9)
1. A method for blood vessel quantitative identification based on deep learning is characterized by comprising the following steps:
constructing a transfer learning network, inputting natural image data into the transfer learning network, and performing network pre-training to obtain a pre-trained recognition network;
carrying out transfer learning training on the pre-trained recognition network by using multi-modal data to obtain a blood vessel quantitative recognition network;
optimizing the blood vessel quantitative recognition network by using perfusion CT image data to obtain a blood vessel quantitative recognition neural network;
and inputting the collected target perfusion CT image into a blood vessel quantitative recognition neural network to obtain a blood vessel quantitative recognition result.
2. The method of claim 1, wherein the pair of natural image data comprises natural image data and scene recognition result.
3. The method of claim 1, wherein the multi-modal data pairs comprise multi-modal image data and vessel region identification data.
4. The method for blood vessel quantitative identification based on deep learning of claim 3, wherein the blood vessel region identification data is: and inputting the multi-modal image data after the vessel region is labeled by an expert or a doctor into the pre-trained recognition network for training to obtain the data.
5. The method of claim 1, wherein the perfusion CT image data pair comprises: perfusion CT image data and vessel region labeling data.
6. The method for blood vessel quantitative identification based on deep learning of claim 5, wherein the blood vessel region labeling data is: and inputting perfusion CT image data after vessel region labeling by a specialist or doctor into the pre-trained recognition network for training to obtain data.
7. An apparatus for vessel quantitative identification based on deep learning, comprising:
the pre-training unit is used for constructing a multi-mode transfer learning network, inputting the natural image data pair into the transfer learning network, and performing network pre-training to obtain a pre-trained recognition network;
the transfer learning training unit is used for carrying out transfer learning training on the pre-trained recognition network through multi-mode data to obtain a blood vessel quantitative recognition network;
the optimization unit is used for optimizing the blood vessel quantitative recognition network through perfusion CT image data to obtain a blood vessel quantitative recognition neural network;
and the result output unit is used for inputting the acquired perfusion CT image into the blood vessel quantitative recognition neural network to obtain a blood vessel quantitative recognition result.
8. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor, when executing the computer program, implements the method for deep learning based vessel quantitative identification as claimed in any one of claims 1-6.
9. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program which, when executed by a processor, causes the processor to carry out the method of deep learning based vessel quantitative identification according to any one of claims 1-6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111089382.6A CN113707312A (en) | 2021-09-16 | 2021-09-16 | Blood vessel quantitative identification method and device based on deep learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111089382.6A CN113707312A (en) | 2021-09-16 | 2021-09-16 | Blood vessel quantitative identification method and device based on deep learning |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113707312A true CN113707312A (en) | 2021-11-26 |
Family
ID=78661210
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111089382.6A Pending CN113707312A (en) | 2021-09-16 | 2021-09-16 | Blood vessel quantitative identification method and device based on deep learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113707312A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116863146A (en) * | 2023-06-09 | 2023-10-10 | 强联智创(北京)科技有限公司 | Method, apparatus and storage medium for extracting hemangio features |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107909566A (en) * | 2017-10-28 | 2018-04-13 | 杭州电子科技大学 | A kind of image-recognizing method of the cutaneum carcinoma melanoma based on deep learning |
CN110459303A (en) * | 2019-06-27 | 2019-11-15 | 浙江工业大学 | Medical imaging abnormal detector based on depth migration |
CN111316290A (en) * | 2017-11-03 | 2020-06-19 | 通用电气公司 | System and method for interactive representation learning migration through deep learning of feature ontologies |
CN111402260A (en) * | 2020-02-17 | 2020-07-10 | 北京深睿博联科技有限责任公司 | Medical image segmentation method, system, terminal and storage medium based on deep learning |
CN111444960A (en) * | 2020-03-26 | 2020-07-24 | 上海交通大学 | Skin disease image classification system based on multi-mode data input |
CN112991346A (en) * | 2021-05-13 | 2021-06-18 | 深圳科亚医疗科技有限公司 | Training method and training system for learning network for medical image analysis |
CN113255794A (en) * | 2021-06-01 | 2021-08-13 | 闽南师范大学 | Medical image classification method based on GoogLeNet network |
-
2021
- 2021-09-16 CN CN202111089382.6A patent/CN113707312A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107909566A (en) * | 2017-10-28 | 2018-04-13 | 杭州电子科技大学 | A kind of image-recognizing method of the cutaneum carcinoma melanoma based on deep learning |
CN111316290A (en) * | 2017-11-03 | 2020-06-19 | 通用电气公司 | System and method for interactive representation learning migration through deep learning of feature ontologies |
CN110459303A (en) * | 2019-06-27 | 2019-11-15 | 浙江工业大学 | Medical imaging abnormal detector based on depth migration |
CN111402260A (en) * | 2020-02-17 | 2020-07-10 | 北京深睿博联科技有限责任公司 | Medical image segmentation method, system, terminal and storage medium based on deep learning |
CN111444960A (en) * | 2020-03-26 | 2020-07-24 | 上海交通大学 | Skin disease image classification system based on multi-mode data input |
CN112991346A (en) * | 2021-05-13 | 2021-06-18 | 深圳科亚医疗科技有限公司 | Training method and training system for learning network for medical image analysis |
CN113255794A (en) * | 2021-06-01 | 2021-08-13 | 闽南师范大学 | Medical image classification method based on GoogLeNet network |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116863146A (en) * | 2023-06-09 | 2023-10-10 | 强联智创(北京)科技有限公司 | Method, apparatus and storage medium for extracting hemangio features |
CN116863146B (en) * | 2023-06-09 | 2024-03-08 | 强联智创(北京)科技有限公司 | Method, apparatus and storage medium for extracting hemangio features |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110232383B (en) | Focus image recognition method and focus image recognition system based on deep learning model | |
CN112102266B (en) | Attention mechanism-based cerebral infarction medical image classification model training method | |
CN110689025B (en) | Image recognition method, device and system and endoscope image recognition method and device | |
Li et al. | MV-RAN: Multiview recurrent aggregation network for echocardiographic sequences segmentation and full cardiac cycle analysis | |
Cui et al. | Multiscale attention guided U-Net architecture for cardiac segmentation in short-axis MRI images | |
CN110475505A (en) | Utilize the automatic segmentation of full convolutional network | |
An et al. | Medical image segmentation algorithm based on multilayer boundary perception-self attention deep learning model | |
CN113205524B (en) | Blood vessel image segmentation method, device and equipment based on U-Net | |
Yan et al. | Cine MRI analysis by deep learning of optical flow: Adding the temporal dimension | |
CN114066904A (en) | Skin lesion image segmentation method and device based on deep learning and storage medium | |
Kumar et al. | Two-stage framework for optic disc segmentation and estimation of cup-to-disc ratio using deep learning technique | |
Yu et al. | Multi-level multi-type self-generated knowledge fusion for cardiac ultrasound segmentation | |
Corbat et al. | Fusion of multiple segmentations of medical images using OV2ASSION and Deep Learning methods: Application to CT-Scans for tumoral kidney | |
CN113707312A (en) | Blood vessel quantitative identification method and device based on deep learning | |
Yu et al. | Dual attention u-net for multi-sequence cardiac MR images segmentation | |
Cao et al. | A novel method for image segmentation: two-stage decoding network with boundary attention | |
CN116703837B (en) | MRI image-based rotator cuff injury intelligent identification method and device | |
Wang et al. | RFPNet: Reorganizing feature pyramid networks for medical image segmentation | |
CN116228690A (en) | Automatic auxiliary diagnosis method for pancreatic cancer and autoimmune pancreatitis based on PET-CT | |
CN111598870B (en) | Method for calculating coronary artery calcification ratio based on convolutional neural network end-to-end reasoning | |
Dong et al. | Transformer-based multi-attention hybrid networks for skin lesion segmentation | |
Zhang et al. | ETUNet: Exploring efficient transformer enhanced UNet for 3D brain tumor segmentation | |
Cai et al. | TSRNet: Tongue image segmentation with global and local refinement | |
CN112634255A (en) | Method and device for establishing brain focus detection model and computer equipment | |
Qi et al. | Exploring generalizable distillation for efficient medical image 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 |