CN111354005A - Full-automatic fetal heart super-image three-blood-vessel segmentation method based on convolutional neural network - Google Patents
Full-automatic fetal heart super-image three-blood-vessel segmentation method based on convolutional neural network Download PDFInfo
- Publication number
- CN111354005A CN111354005A CN202010128494.7A CN202010128494A CN111354005A CN 111354005 A CN111354005 A CN 111354005A CN 202010128494 A CN202010128494 A CN 202010128494A CN 111354005 A CN111354005 A CN 111354005A
- Authority
- CN
- China
- Prior art keywords
- neural network
- convolutional neural
- image
- blood
- heart
- 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
- 238000013527 convolutional neural network Methods 0.000 title claims abstract description 62
- 230000011218 segmentation Effects 0.000 title claims abstract description 49
- 238000000034 method Methods 0.000 title claims abstract description 42
- 210000002458 fetal heart Anatomy 0.000 title claims abstract description 17
- 238000012549 training Methods 0.000 claims abstract description 48
- 210000003754 fetus Anatomy 0.000 claims abstract description 26
- 210000004204 blood vessel Anatomy 0.000 claims abstract description 14
- 238000012545 processing Methods 0.000 claims abstract description 10
- 230000001605 fetal effect Effects 0.000 claims abstract description 8
- 210000000709 aorta Anatomy 0.000 claims description 10
- 210000001147 pulmonary artery Anatomy 0.000 claims description 10
- 210000002620 vena cava superior Anatomy 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 8
- 238000002604 ultrasonography Methods 0.000 claims description 8
- 238000005070 sampling Methods 0.000 claims description 7
- 230000003321 amplification Effects 0.000 claims description 5
- 238000013434 data augmentation Methods 0.000 claims description 5
- 238000003199 nucleic acid amplification method Methods 0.000 claims description 5
- 238000005457 optimization Methods 0.000 claims description 5
- 230000000747 cardiac effect Effects 0.000 claims description 4
- 230000009466 transformation Effects 0.000 claims description 4
- 238000006073 displacement reaction Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 claims description 3
- 238000011176 pooling Methods 0.000 claims description 3
- 238000002360 preparation method Methods 0.000 claims description 3
- 208000002330 Congenital Heart Defects Diseases 0.000 abstract description 4
- 238000012216 screening Methods 0.000 abstract description 4
- 238000005516 engineering process Methods 0.000 abstract description 3
- 208000028831 congenital heart disease Diseases 0.000 abstract description 2
- 238000009795 derivation Methods 0.000 abstract description 2
- 201000010099 disease Diseases 0.000 description 3
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 206010010356 Congenital anomaly Diseases 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000003190 augmentative effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 208000018695 congenital heart malformation Diseases 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000005831 heart abnormality Effects 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 230000036244 malformation Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
- 238000012285 ultrasound imaging Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- 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
- G06T2207/10132—Ultrasound image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30044—Fetus; Embryo
-
- 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
- G06T2207/30048—Heart; Cardiac
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30101—Blood vessel; Artery; Vein; Vascular
Abstract
The invention relates to a medical image processing technology, and aims to provide a full-automatic fetal heart super-image three-blood-vessel segmentation method based on a convolutional neural network. The method comprises the following steps: (1) carrying out standard pretreatment and calibration on the ultrasonic image of the three blood vessel sections of the heart of the fetus, and preparing a training set; (2) constructing and training a convolutional neural network; (3) and carrying out data processing on the new ultrasonic image of the three blood vessel sections of the heart of the fetus by using the trained convolutional neural network to obtain a segmentation result. The invention carries out three-vessel segmentation on the ultrasonic image of the heart of the fetus by means of the convolutional neural network, and can automatically realize the derivation of the segmentation result. The segmentation result can be used for screening fetal congenital heart disease, the accuracy rate of the three-blood-vessel segmentation reaches the average intersection ratio not less than 80%, and the acceptable requirement of clinical application is met.
Description
Technical Field
The invention relates to the field of medical image processing, in particular to a full-automatic fetal heart super image three-blood-vessel segmentation method based on a convolutional neural network.
Background
Ultrasound imaging can be used for screening of fetal fatal malformations, particularly in screening of fetal congenital heart malformations. The three-vessel section of the ultrasonic image of the fetal heart is one of the main standard sections for examining the fetal heart, the three-vessel section shows the normal sequence of three vessels including the superior vena cava, the aorta and the pulmonary artery and the three vessels, and if the three-vessel sequence is abnormal, the congenital heart abnormality of the fetus is indicated. Due to the characteristics of more noise points, blurring and the like of the ultrasonic image, the traditional image segmentation method cannot obtain a satisfactory result on the ultrasonic image of the heart of the fetus. At present, the ultrasonic image examination of the heart of the fetus mainly passes through the manual examination of an ultrasonic expert, and at present, a full-automatic segmentation method is not available on the three-blood-vessel section of the ultrasonic image of the heart of the fetus. The manual interpretation has the defects of large workload, long measuring period, strong subjectivity and the like. Therefore, a well established three-blood-vessel automatic segmentation method is urgently needed, and a full-automatic fetal heart ultrasonic image three-blood-vessel segmentation method capable of meeting clinical requirements is provided, which is very necessary for clinical medical application.
Disclosure of Invention
The invention aims to solve the technical problem of overcoming the defects in the prior art and provides a full-automatic fetal heart super image three-blood-vessel segmentation method based on a convolutional neural network.
In order to solve the technical problem, the solution of the invention is as follows:
the full-automatic fetal heart super image three-blood-vessel segmentation method based on the convolutional neural network comprises the following steps:
(1) carrying out standard pretreatment and calibration on the ultrasonic image of the three blood vessel sections of the heart of the fetus, and preparing a training set;
the method specifically comprises the following steps:
(1.1) collecting the ultrasonic image of the section of the three blood vessels of the heart of the fetus from an ultrasonic image database of a hospital;
(1.2) carrying out image standardized sampling on the collected ultrasonic image, wherein the image size is 1024 × 768;
(1.3) calibrating the preprocessed ultrasonic image, and drawing corresponding superior vena cava, aorta and pulmonary artery regions according to a conventional medical image identification rule in a manual mode;
(2) constructing and training a convolutional neural network;
the method specifically comprises the following steps:
(2.1) randomly selecting 90% from the calibrated ultrasonic images as a training set, and performing data augmentation on the training set;
(2.2) constructing a convolutional neural network for learning training; the convolutional neural network is alternately realized by a plurality of convolutional layers, a cavity convolutional layer, a feature extraction block and a pooling layer;
(2.3) inputting the training set subjected to data amplification in the step (2.1) into the convolutional neural network in the step (2.2), and training parameters in the convolutional neural network; after a plurality of times of training, obtaining a learned network weight parameter;
(3) carrying out data processing on the new ultrasonic image of the three blood vessel sections of the heart of the fetus by using the trained convolutional neural network to obtain a segmentation result;
the method specifically comprises the following steps:
(3.1) carrying out image standardized sampling on the new ultrasonic image to obtain an image with the size of 1024 × 768;
(3.2) inputting the image data obtained in the step (3.1) into a trained convolutional neural network, applying network weight parameters for training and learning of a training set through iteration of the convolutional neural network, and outputting a point-by-point classification result of the image;
and (3.3) obtaining segmentation results of three corresponding regions of the superior vena cava, the aorta and the pulmonary artery according to the point-by-point classification result of the image obtained in the step (3.2).
In the invention, when the data of the training set is augmented in the step (2.1), the data volume is increased to 5 times of the original data volume; the adopted data augmentation method is one or the combination of more than two of the following items: the scale transformation of the foreground image in a set range, the displacement in a set range, or the random rotation in a set angle range.
In the invention, when the convolutional neural network is used for learning and training, the loss function value is reduced through an optimization algorithm.
The invention further provides a full-automatic fetal heart ultrasonic image three-blood-vessel segmentation device based on the convolutional neural network, which comprises the following components:
the standardized preprocessing module is used for carrying out standardized preprocessing and calibration on the ultrasonic image of the three blood vessel sections of the heart of the fetus to finish the preparation of a training set;
the convolutional neural network training module is used for training a convolutional neural network;
and the three-blood-vessel segmentation module is used for carrying out data processing on the new ultrasonic image of the three blood-vessel section of the heart of the fetus by using the trained convolutional neural network to obtain segmentation results of three areas, namely the superior vena cava, the aorta and the pulmonary artery.
The invention further provides a full-automatic fetal heart ultrasonic image three-blood-vessel segmentation device based on the convolutional neural network, which comprises a memory and a processor;
the memory for storing a computer program;
the processor, when executing the computer program, is capable of implementing the fully automatic fetal cardiac ultrasound image three-vessel segmentation method based on the convolutional neural network as claimed in any one of claims 1 to 3.
The present invention further provides a computer readable storage medium, on which a computer program is stored, which, when being executed by a processor, is capable of implementing the fully automatic fetal heart ultrasound image three-vessel segmentation method based on convolutional neural network as set forth in any one of claims 1 to 3.
Compared with the prior art, the invention has the beneficial effects that:
1. the invention carries out three-vessel segmentation on the ultrasonic image of the heart of the fetus by means of the convolutional neural network, and can automatically realize the derivation of the segmentation result.
2. The segmentation result of the invention can be used for screening fetal congenital heart disease. The applicant compared the results obtained by segmentation with the method of the invention with the results of examination of each sample, using a sufficient number of foetal cardiac ultrasound image samples meeting the statistical requirements, obtained at several hospitals. Statistical data show that the accuracy of the three-vessel segmentation of the invention reaches the average intersection ratio (MIOU) of not less than 80%, and the three-vessel segmentation meets the acceptable requirements of clinical application.
Drawings
FIG. 1 is a schematic flow diagram of the present invention;
FIG. 2 is an ultrasound raw image for use in the present invention;
FIG. 3 is a schematic diagram of a process for normalization preprocessing and calibration of an ultrasound image;
FIG. 4 is a flowchart of outputting a three-vessel segmentation result by using a convolutional neural network according to the present invention.
Detailed Description
It should be noted that the present invention relates to a database technology, which is an application of a computer technology in the field of medical image processing. In the implementation process of the invention, the application of a plurality of software functional modules is involved. The applicant believes that it is fully possible for one skilled in the art to utilize the software programming skills in his or her own practice to implement the invention, as well as to properly understand the principles and objectives of the invention, in conjunction with the prior art, after a perusal of this application. The aforementioned software functional modules include but are not limited to: the standardized preprocessing module, the convolutional neural network training module, the three-vessel segmentation module and the like belong to the scope mentioned in the present application document, and the applicant does not list the modules one by one.
The memory, the processor and the computer readable storage medium are all hardware devices in the computer industry, and the invention has no special requirement on the hardware devices. Except for the specifically described contents, the construction method and the training method of the convolutional neural network can adopt the conventional mode in the field, so the details are not repeated.
The invention is described in further detail below with reference to the following detailed description and accompanying drawings:
the invention discloses a full-automatic fetal heart ultrasonic image three-blood-vessel segmentation method based on a convolutional neural network, which specifically comprises the following processes of:
firstly, preparing a training set; the method specifically comprises the following steps:
a, collecting 5000 cases of three-vessel section ultrasonic images from an ultrasonic database of a hospital, carrying out image standardized sampling on the ultrasonic images, wherein the size of the images is 1024 × 768, then carrying out calibration, and drawing out corresponding superior vena cava, aorta and pulmonary artery areas.
Secondly, constructing and training a convolutional neural network; the method specifically comprises the following steps:
and B: and randomly selecting 90% of data from the preprocessed and calibrated ultrasonic data as a training set, and performing data amplification on the training set to increase the data volume to 5 times of the original data volume. The data augmentation methods adopted mainly include but are not limited to: the scale transformation of the foreground image in a set range, the displacement of a set degree, the random rotation in a set angle range and the organic combination of the three transformations;
and C: constructing a convolutional neural network for learning training, wherein the convolutional neural network is alternately realized by a plurality of convolutional layers, a cavity convolutional layer, a feature extraction block and a pooling layer;
step D: inputting the training set subjected to data amplification into a convolutional neural network, and training parameters in the convolutional neural network; after a plurality of times of training, obtaining a learned network weight parameter; and the loss function value in the learning training process is reduced through an optimization algorithm. (Note: optimization algorithm is a mathematical term referring to the algorithm used to solve the optimization problem, typically solving for local extrema under specific conditions.)
And thirdly, processing the ultrasonic image data of the three-blood-vessel section of the heart of the fetus by using the trained convolutional neural network to obtain a three-blood-vessel segmentation result.
The method specifically comprises the following steps:
e, selecting an ultrasonic image of a three-vessel section of the heart of the fetus to be screened, and carrying out standardized sampling on the ultrasonic image according to the method in the step A, wherein the sampling size is 1024 × 768;
step F: e, taking the image data in the step E as the input of the trained convolutional neural network, applying network weight parameters obtained by training and learning the training set through the iteration of the convolutional neural network, and outputting corresponding image point-by-point classification results;
step G: and F, obtaining corresponding segmentation results of the superior vena cava, the aorta and the pulmonary artery region according to the point-by-point classification result of the image obtained in the step F.
Based on the full-automatic fetal heart ultrasonic image three-blood-vessel segmentation method, the invention also provides a corresponding device and a computer readable storage medium, and the method specifically comprises the following steps:
a full-automatic fetus heart ultrasonic image three-blood-vessel segmentation device based on a convolutional neural network comprises:
the standardized preprocessing module is used for carrying out standardized preprocessing and calibration on the ultrasonic image of the three blood vessel sections of the heart of the fetus to finish the preparation of a training set; the convolutional neural network training module is used for training a convolutional neural network; and the three-blood-vessel segmentation module is used for carrying out data processing on the new ultrasonic image of the three blood-vessel section of the heart of the fetus by using the trained convolutional neural network to obtain segmentation results of three areas, namely the superior vena cava, the aorta and the pulmonary artery.
A full-automatic fetus heart ultrasonic image three-blood-vessel segmentation device based on a convolutional neural network comprises a memory and a processor; the memory for storing a computer program; the processor is used for realizing the fully-automatic fetal heart ultrasonic image three-blood-vessel segmentation method based on the convolutional neural network when executing the computer program.
A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, is capable of implementing the fully automatic fetal cardiac ultrasound image three-vessel segmentation method based on a convolutional neural network as described above.
In addition, the applicant needs to emphasize that the final result obtained by the present invention can only be used as reference test data of fetal congenital heart diseases in medical practice, but cannot be directly used for judging whether the subject has diseases or not, and even cannot be used as a disease treatment means for the subject. Therefore, the technical scheme of the invention does not have the purpose of diagnosing or treating diseases per se.
Finally, it should be noted that the above-mentioned list is only a specific embodiment of the present invention. It is obvious that the present invention is not limited to the above embodiments, but many variations are possible. All modifications which can be derived or suggested by a person skilled in the art from the disclosure of the present invention are to be considered within the scope of the invention.
Claims (6)
1. A full-automatic fetal heart super image three-blood-vessel segmentation method based on a convolutional neural network is characterized by comprising the following steps:
(1) carrying out standard pretreatment and calibration on the ultrasonic image of the three blood vessel sections of the heart of the fetus, and preparing a training set;
the method specifically comprises the following steps:
(1.1) collecting the ultrasonic image of the section of the three blood vessels of the heart of the fetus from an ultrasonic image database of a hospital;
(1.2) carrying out image standardized sampling on the collected ultrasonic image, wherein the image size is 1024 × 768;
(1.3) calibrating the preprocessed ultrasonic image, and drawing corresponding superior vena cava, aorta and pulmonary artery regions according to a conventional medical image identification rule in a manual mode;
(2) constructing and training a convolutional neural network;
the method specifically comprises the following steps:
(2.1) randomly selecting 90% from the calibrated ultrasonic images as a training set, and performing data augmentation on the training set;
(2.2) constructing a convolutional neural network for learning training; the convolutional neural network is alternately realized by a plurality of convolutional layers, a cavity convolutional layer, a feature extraction block and a pooling layer;
(2.3) inputting the training set subjected to data amplification in the step (2.1) into the convolutional neural network in the step (2.2), and training parameters in the convolutional neural network; after a plurality of times of training, obtaining a learned network weight parameter;
(3) carrying out data processing on the new ultrasonic image of the three blood vessel sections of the heart of the fetus by using the trained convolutional neural network to obtain a segmentation result;
the method specifically comprises the following steps:
(3.1) carrying out image standardized sampling on the new ultrasonic image to obtain an image with the size of 1024 × 768;
(3.2) inputting the image data obtained in the step (3.1) into a trained convolutional neural network, applying network weight parameters for training and learning of a training set through iteration of the convolutional neural network, and outputting a point-by-point classification result of the image;
and (3.3) obtaining segmentation results of three corresponding regions of the superior vena cava, the aorta and the pulmonary artery according to the point-by-point classification result of the image obtained in the step (3.2).
2. The method according to claim 1, wherein in the step (2.1), when data amplification is performed on the training set, the data volume is increased to 5 times of the original data volume; the adopted data augmentation method is one or the combination of more than two of the following items: the scale transformation of the foreground image in a set range, the displacement in a set range, or the random rotation in a set angle range.
3. The method of claim 1, wherein the loss function values are reduced by an optimization algorithm when performing the learning training using a convolutional neural network.
4. A full-automatic fetus heart ultrasonic image three-blood-vessel segmentation device based on a convolutional neural network is characterized by comprising:
the standardized preprocessing module is used for carrying out standardized preprocessing and calibration on the ultrasonic image of the three blood vessel sections of the heart of the fetus to finish the preparation of a training set;
the convolutional neural network training module is used for training a convolutional neural network;
and the three-blood-vessel segmentation module is used for carrying out data processing on the new ultrasonic image of the three blood-vessel section of the heart of the fetus by using the trained convolutional neural network to obtain segmentation results of three areas, namely the superior vena cava, the aorta and the pulmonary artery.
5. A full-automatic fetus heart ultrasonic image three-blood-vessel segmentation device based on a convolutional neural network is characterized by comprising a memory and a processor;
the memory for storing a computer program;
the processor, when executing the computer program, is capable of implementing the fully automatic fetal cardiac ultrasound image three-vessel segmentation method based on the convolutional neural network as claimed in any one of claims 1 to 3.
6. A computer-readable storage medium, wherein the storage medium has stored thereon a computer program, which when executed by a processor, is capable of implementing the fully automatic fetal heart ultrasound image three-vessel segmentation method based on convolutional neural network as claimed in any one of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010128494.7A CN111354005A (en) | 2020-02-28 | 2020-02-28 | Full-automatic fetal heart super-image three-blood-vessel segmentation method based on convolutional neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010128494.7A CN111354005A (en) | 2020-02-28 | 2020-02-28 | Full-automatic fetal heart super-image three-blood-vessel segmentation method based on convolutional neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111354005A true CN111354005A (en) | 2020-06-30 |
Family
ID=71197161
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010128494.7A Pending CN111354005A (en) | 2020-02-28 | 2020-02-28 | Full-automatic fetal heart super-image three-blood-vessel segmentation method based on convolutional neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111354005A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112465786A (en) * | 2020-12-01 | 2021-03-09 | 平安科技（深圳）有限公司 | Model training method, data processing method, device, client and storage medium |
| CN116912236A (en) * | 2023-09-08 | 2023-10-20 | 首都医科大学附属北京妇产医院 | Method, system and storable medium for predicting fetal congenital heart disease risk based on artificial intelligence |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20170113251A (en) * | 2016-03-24 | 2017-10-12 | 재단법인 아산사회복지재단 | Method and device for automatic inner and outer vessel wall segmentation in intravascular ultrasound images using deep learning |
| 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 |
| WO2020001217A1 (en) * | 2018-06-27 | 2020-01-02 | 东南大学 | Segmentation method for dissected aorta in ct image based on convolutional neural network |
-
2020
- 2020-02-28 CN CN202010128494.7A patent/CN111354005A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20170113251A (en) * | 2016-03-24 | 2017-10-12 | 재단법인 아산사회복지재단 | Method and device for automatic inner and outer vessel wall segmentation in intravascular ultrasound images using deep learning |
| WO2020001217A1 (en) * | 2018-06-27 | 2020-01-02 | 东南大学 | Segmentation method for dissected aorta in ct image based on convolutional neural network |
| 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 |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112465786A (en) * | 2020-12-01 | 2021-03-09 | 平安科技（深圳）有限公司 | Model training method, data processing method, device, client and storage medium |
| CN116912236A (en) * | 2023-09-08 | 2023-10-20 | 首都医科大学附属北京妇产医院 | Method, system and storable medium for predicting fetal congenital heart disease risk based on artificial intelligence |
| CN116912236B (en) * | 2023-09-08 | 2023-12-26 | 首都医科大学附属北京妇产医院 | Method, system and storable medium for predicting fetal congenital heart disease risk based on artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110706246B (en) | Blood vessel image segmentation method and device, electronic equipment and storage medium | |
| CN108052977B (en) | Mammary gland molybdenum target image deep learning classification method based on lightweight neural network | |
| JP5926728B2 (en) | Visualization adapted for direct use by physicians | |
| Badsha et al. | A new blood vessel extraction technique using edge enhancement and object classification | |
| CN112070067B (en) | Scatter diagram classification method and device for photoplethysmograph signals | |
| CN112381164B (en) | Ultrasound image classification method and device based on multi-branch attention mechanism | |
| TW202008211A (en) | Method and electronic apparatus for image processing | |
| CN111354005A (en) | Full-automatic fetal heart super-image three-blood-vessel segmentation method based on convolutional neural network | |
| CN110310723A (en) | Bone image processing method, electronic equipment and storage medium | |
| Wang et al. | Automatic real-time CNN-based neonatal brain ventricles segmentation | |
| CN114119637A (en) | Brain white matter high signal segmentation method based on multi-scale fusion and split attention | |
| CN115147600A (en) | GBM multi-mode MR image segmentation method based on classifier weight converter | |
| CN110555846A (en) | full-automatic bone age assessment method based on convolutional neural network | |
| CN111584072A (en) | Neural network model training method suitable for small samples | |
| CN110930373A (en) | Pneumonia recognition device based on neural network | |
| CN114066846A (en) | CTP non-acute occlusion ischemia assessment method and system based on deep learning | |
| Moses et al. | Automatic segmentation and analysis of the main pulmonary artery on standard post-contrast CT studies using iterative erosion and dilation | |
| US7558427B2 (en) | Method for analyzing image data | |
| CN115810018A (en) | Method and system for optimizing segmentation results of blood vessel tree and coronary artery tree of CT image | |
| CN113222985B (en) | Image processing method, image processing device, computer equipment and medium | |
| CN115222674A (en) | Detection device for intracranial aneurysm rupture risk based on multi-dimensional feature fusion | |
| CN114429468A (en) | Bone age measuring method, bone age measuring system, electronic device and computer-readable storage medium | |
| Taş et al. | Detection of retinal diseases from ophthalmological images based on convolutional neural network architecture. | |
| Salehi et al. | Investigation and simulation of different medical image processing algorithms to improve image quality using simulink matlab | |
| CN116759042B (en) | System and method for generating anti-facts medical data based on annular consistency |
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 |