CN110060315A - A kind of image motion artifact eliminating method and system based on artificial intelligence - Google Patents
A kind of image motion artifact eliminating method and system based on artificial intelligence Download PDFInfo
- Publication number
- CN110060315A CN110060315A CN201910325563.0A CN201910325563A CN110060315A CN 110060315 A CN110060315 A CN 110060315A CN 201910325563 A CN201910325563 A CN 201910325563A CN 110060315 A CN110060315 A CN 110060315A
- Authority
- CN
- China
- Prior art keywords
- image
- neural network
- artificial intelligence
- motion
- motion artifact
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 52
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000013528 artificial neural network Methods 0.000 claims abstract description 58
- 238000012549 training Methods 0.000 claims abstract description 27
- 230000001537 neural effect Effects 0.000 claims abstract description 10
- 238000012545 processing Methods 0.000 claims description 20
- 230000006870 function Effects 0.000 claims description 19
- 238000011478 gradient descent method Methods 0.000 claims description 8
- 238000005457 optimization Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 abstract description 6
- 238000013135 deep learning Methods 0.000 abstract description 5
- 230000008569 process Effects 0.000 description 6
- 230000008030 elimination Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 238000013519 translation Methods 0.000 description 3
- 238000002591 computed tomography Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 210000004218 nerve net Anatomy 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000003325 tomography Methods 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013170 computed tomography imaging Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000002059 diagnostic imaging Methods 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000005764 inhibitory process Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000009659 non-destructive testing Methods 0.000 description 1
- 230000010363 phase shift Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G06T5/73—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/003—Reconstruction from projections, e.g. tomography
- G06T11/008—Specific post-processing after tomographic reconstruction, e.g. voxelisation, metal artifact correction
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20172—Image enhancement details
- G06T2207/20201—Motion blur correction
-
- 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
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention discloses a kind of image motion artifact eliminating method and system based on artificial intelligence, shown method is the following steps are included: handle target image to obtain the image with motion artifacts, and construct image data base；Neural network is constructed according to image data base, and carries out neural metwork training；The motion artifacts of image to be processed are eliminated using trained neural network and export processed image.Due to the deep learning method based on artificial intelligence, by building data training Network Recognition and such image artifacts are reduced or eliminated, do not have particular/special requirement to the track of movement, and flexibly, calculation amount is small, and robustness is preferable.
Description
Technical field
The present invention relates to medical imaging technology field more particularly to a kind of image motion artifacts based on artificial intelligence
Removing method and system.
Background technique
CT (computer dislocation scanning and imaging system) utilizes X-ray scanning object, obtains data for projection, and pass through tomography weight
Algorithm is built to handle these data for projection, the tomography and three-dimensional density information of object is obtained, achievees the purpose that non-destructive testing.Cause
The complexity of Imaging physics process will cope with various image artifacts to the shadow of final image quality state-of-the-art CT system
It rings.Motion artifacts are particularly evident in performance wherein.
In the prior art, the movement of patient is inevitable in clinical application, and there are many kinds of modes, for fortune
Dynamic artifact, is the method based on estimation mostly at present, but this method needs the scanning of redundancy to estimate the fortune of object
Dynamic rail mark, and because be the method based on iteration, calculation amount is very big, the result of estimation also inadequate robust.
Therefore, the existing technology needs to be improved and developed.
Summary of the invention
The technical problem to be solved in the present invention is that in view of the above drawbacks of the prior art, providing a kind of based on artificial intelligence
The image motion artifact eliminating method and system of energy, it is intended to which the calculation amount for solving alternative manner estimation in the prior art causes greatly very much
The problem of poor robustness.
The technical proposal for solving the technical problem of the invention is as follows:
A kind of image motion artifact eliminating method based on artificial intelligence, wherein comprising steps of
Target image is handled to obtain the image with motion artifacts, and constructs image data base；
Neural network is constructed according to image data base, and carries out neural metwork training；
The motion artifacts of image to be processed are eliminated using trained neural network and export processed image.
The image motion artifact eliminating method based on artificial intelligence, wherein described that target image is handled to obtain
Image with motion artifacts, and construct image database steps and specifically include:
Two dimension is carried out to target image or three-dimensional Fourier transform forms Fourier space；
After increasing phase offset in the Fourier space of each angle, carries out inverse-Fourier transform and obtain with movement puppet
The image of shadow；
The corresponding target image composition figure of image with motion artifacts is opposite, and form image data base.
The image motion artifact eliminating method based on artificial intelligence, wherein described to be constructed according to image data base
Neural network, and carry out neural metwork training step and specifically include:
The neural network based on convolution is constructed according to image data base；
Using gradient descent method, according to the parameter of objective function optimization neural network, and complete to train.
The image motion artifact eliminating method based on artificial intelligence, wherein the objective function are as follows:
Wherein, Img is the image crossed by Processing with Neural Network, ImgtargFor target image, k is positive integer, ImgkIt is
The k images crossed by Processing with Neural Network, Imgtarg,kFor kth target image, Σ is sum operation, and Loss () is mesh
Scalar functions.
The image motion artifact eliminating method based on artificial intelligence, wherein in the nerve net based on convolution
In network, the label of background and various artifact regions is exported when the image that output is crossed by Processing with Neural Network.
A kind of image motion artifact based on artificial intelligence eliminates system, wherein include: processor, and with the place
The memory of device connection is managed,
The memory is stored with image motion artifact based on artificial intelligence and eliminates program, described based on artificial intelligence
Image motion artifact is eliminated when program is executed by the processor and is performed the steps of
Target image is handled to obtain the image with motion artifacts, and constructs image data base；
Neural network is constructed according to image data base, and carries out neural metwork training；
The motion artifacts of image to be processed are eliminated using trained neural network and export processed image.
The image motion artifact based on artificial intelligence eliminates system, wherein the image based on artificial intelligence
When motion artifacts elimination program is executed by the processor, also perform the steps of
Two dimension is carried out to target image or three-dimensional Fourier transform forms Fourier space；
After increasing phase offset in the Fourier space of each angle, carries out inverse-Fourier transform and obtain with movement puppet
The image of shadow；
The corresponding target image composition figure of image with motion artifacts is opposite, and form image data base.
The image motion artifact based on artificial intelligence eliminates system, wherein the image based on artificial intelligence
When motion artifacts elimination program is executed by the processor, also perform the steps of
The neural network based on convolution is constructed according to image data base；
Using gradient descent method, according to the parameter of objective function optimization neural network, and complete to train.
The image motion artifact based on artificial intelligence eliminates system, wherein the objective function are as follows:
Wherein, Img is the image crossed by Processing with Neural Network, ImgtargFor target image, k is positive integer, ImgkIt is
The k images crossed by Processing with Neural Network, Imgtarg,kFor kth target image, Σ is sum operation, and Loss () is mesh
Scalar functions.
The image motion artifact based on artificial intelligence eliminates system, wherein in the nerve net based on convolution
In network, the label of background and various artifact regions is exported in the image that output is crossed by Processing with Neural Network.
The utility model has the advantages that by building data training Network Recognition and subtracting due to the deep learning method based on artificial intelligence
Less or such image artifacts are eliminated, there is no particular/special requirement to the track of movement, and flexibly, calculation amount is small, and robustness is preferable.
Detailed description of the invention
Fig. 1 is the image motion artifact eliminating method First Principle figure in the present invention based on artificial intelligence.
Fig. 2 is the second schematic diagram of image motion artifact eliminating method in the present invention based on artificial intelligence.
Fig. 3 is the image motion artifact eliminating method third schematic diagram in the present invention based on artificial intelligence.
Fig. 4 is the image motion artifact eliminating method flow chart in the present invention based on artificial intelligence.
Fig. 5 is the schematic diagram of the function of the image motion artifact elimination system in the present invention based on artificial intelligence.
Specific embodiment
To make the objectives, technical solutions, and advantages of the present invention clearer and more explicit, right as follows in conjunction with drawings and embodiments
The present invention is further described.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and do not have to
It is of the invention in limiting.
Please refer to Fig. 1-Fig. 4, the present invention provides a kind of image motion artifact eliminating method based on artificial intelligence
Some embodiments.
As shown in figure 4, a kind of image motion artifact eliminating method based on artificial intelligence of the invention, including following step
It is rapid:
Step S100, target image is handled to obtain the image with motion artifacts, and constructs image data base.
Specifically, the step S100 specifically includes the following steps:
S110, two dimension or three-dimensional Fourier transform formation Fourier space are carried out to target image.
S120, it after increasing phase offset in the Fourier space of each angle, carries out inverse-Fourier transform and is had
The image of motion artifacts.
S130, the corresponding target image composition of the image with motion artifacts is schemed relatively, and forms image data
Library.
The sample of the database of training neural network generally requires and pairs of provides input picture (comprising motion artifacts
Image), the image of target image (ideal image, without artifact).But for motion artifacts, input picture and target figure
Difference as between, which mainly misplaces, causes difference, and the difference of artifact bring is more had ignored by neural network.And facing
Also it is difficult that acquisition has movement of patient and the data of movement of patient do not generate pairs of training data in bed data.Here it proposes
A kind of method for relying solely on patient image data and capable of generating pairs of training data.According to the property of Fourier transform, one
The Fourier transform of the Fourier transform of a image and this same image after translation, differ only by one and translation away from
It is specific as follows from related phase:
Fourier transform and CT scan have very close connection again, because the data of each angle scanning are equivalent to
The data of same angle in Fourier space.
The two properties are combined, propose the method for skimulated motion object of which movement artifact here.As shown in Figure 2.Input
The image not moved.By two-dimensional FFT transform, increase phase offset (such as Fig. 1 in the Fourier space of each angle
It is shown), it then carries out inverse-Fourier transform again on this basis and has just obtained the image with motion artifacts.In terms of result this
The image encountered in a image and clinic is very similar.Which produces the image pair of training below, pairs of matched image
It can be generated from clinical image data or emulation data.In pairs of matched image, first group and second group respectively be can be
Piece image or one group of image.Smart network can be trained by a certain number of matched images in pairs.First group
It is the image with motion artifacts that simulation generates, second group is the ideal image of quality, i.e. piece image is through artificial intelligence
The image of network conversion and acquisition.First group picture seems to be produced by the method for random phase translation come Fast simulation motion artifacts
Raw includes the image of motion artifacts.Second group of image is then the method that is filtered to the phase shift used in first group to produce
Image after raw artifact inhibition.
Step S200, neural network is constructed according to image data base, and carries out neural metwork training.
Here it is illustrated by taking two-dimensional image data as an example.
1, training data: the scan data of the not motion artifacts obtained from normal CT scan is collected, it is random by phase
The Fourier's series and inversion process of offset obtain the image of skimulated motion artifact.As shown in Fig. 2, original image and simulation fortune
The image of dynamic artifact is formed one group of basic training data.
2, planned network structure: the structure of network can be much to select, here mainly using convolutional neural networks as base
The network of plinth, for example use U-net network.
3, network training: network and data put in order after just using the frame of general deep learning such as
Tensorflow is trained.Loss function optimizes weight coefficient using gradient descent method with regard to as shown previously.
Specifically, the step S200 specifically comprises the following steps:
Step S210, the neural network based on convolution is constructed according to image data base.
Specifically, the selected model of training is mainly based upon the neural network of convolution, such as improved ResNet,
Unet.With the difference of initial network, network here after input picture, in addition to output processing after image it
Also to export the label of background and various artifact regions simultaneously outside.That is in the neural network based on convolution, work as output
The label of background and various artifact regions is exported when the image crossed by Processing with Neural Network.
Step S220, it using gradient descent method, according to the parameter of objective function optimization neural network, and completes to train.
Specifically, the objective function are as follows:
Wherein, Img is the image crossed by Processing with Neural Network, ImgtargFor target image, k is positive integer, ImgkIt is
The k images crossed by Processing with Neural Network, Imgtarg,kFor kth target image, Σ is sum operation, and Loss () is mesh
Scalar functions.
After setting these parameters, it is directed to objective function using gradient descent method, the parameter of neural network is optimized.
After the completion of training, the parameter of entire model is retained, for using later.Based in training process smart network at
Ripe degree, by a certain number of test images occurred in pairs checked except tranining database, to determine network training result
Quality.When the piece image in these test images is successfully converted into new images, and the new figure by smart network
As this can be showed by the standard set in training process in image when the quality of the second width image (target image), the training
Process can terminate.
Step S300, the motion artifacts of image to be processed are eliminated using trained neural network and exports processed figure
Picture.
Specifically, as shown in figure 3, eliminating the motion artifacts of image to be processed using trained neural network and exporting
Handle image.
It is worth noting that in the physical process of X ray CT imaging, because of the limitation of scanning speed, because of the fortune of patient
It is dynamic to avoid completely, image also often can there are the artifacts moved to a certain degree, to influence diagnostic result.Such artifact often with
The form of striped embodies in the picture.The invention proposes the deep learning methods based on artificial intelligence, pass through building data instruction
Practice Network Recognition and simultaneously reduce or eliminate such image artifacts, so as to improve image, preferably shows the internal structure of illuminated object
And tissue.
Method of the invention does not have particular/special requirement to the track of movement, and flexibly, calculation amount is small, and robustness is preferable.Only
There are enough training datas, neural network can formidably distinguish corresponding artifact enough, can cover wider system
System.A little self-adapting estimation and processing for changing just can reach to similar artifact accordingly are added in data.
The present invention also provides a kind of, and the image motion artifact based on artificial intelligence eliminates the preferred embodiment of system:
As shown in figure 5, a kind of image motion artifact based on artificial intelligence of the embodiment of the present invention eliminates system, comprising:
Processor 10, and the memory 20 being connect with the processor 10,
The memory 20 is stored with the image motion artifact based on artificial intelligence and eliminates program, described to be based on artificial intelligence
Image motion artifact eliminate program and perform the steps of when being executed by the processor 10
Target image is handled to obtain the image with motion artifacts, and constructs image data base；
Neural network is constructed according to image data base, and carries out neural metwork training；
The motion artifacts of image to be processed are eliminated using trained neural network and export processed image, it is specific as above
It is described.
When the image motion artifact elimination program based on artificial intelligence is executed by the processor 10, also realize following
Step:
Two dimension is carried out to target image or three-dimensional Fourier transform forms Fourier space；
After increasing phase offset in the Fourier space of each angle, carries out inverse-Fourier transform and obtain with movement puppet
The image of shadow；
The corresponding target image composition figure of image with motion artifacts is opposite, and image data base is formed, have
Body is as described above.
When the image motion artifact elimination program based on artificial intelligence is executed by the processor 10, also realize following
Step:
The neural network based on convolution is constructed according to image data base；
It using gradient descent method, according to the parameter of objective function optimization neural network, and completes to train, specific institute as above
It states.
The objective function are as follows:
Wherein, Img is the image crossed by Processing with Neural Network, ImgtargFor target image, k is positive integer, ImgkIt is
The k images crossed by Processing with Neural Network, Imgtarg,kFor kth target image, Σ is sum operation, and Loss () is mesh
Scalar functions, as detailed above.
In the neural network based on convolution, in the image that output is crossed by Processing with Neural Network output background and
The label of various artifact regions.
In conclusion a kind of image motion artifact eliminating method and system based on artificial intelligence provided by the present invention,
Shown method is the following steps are included: handle target image to obtain the image with motion artifacts, and construct image data base；Root
Neural network is constructed according to image data base, and carries out neural metwork training；Figure to be processed is eliminated using trained neural network
The motion artifacts of picture simultaneously export processed image.Due to the deep learning method based on artificial intelligence, pass through building data training
Network Recognition simultaneously reduces or eliminates such image artifacts, does not have particular/special requirement to the track of movement, and flexibly, calculation amount is small,
Robustness is preferable.
It should be understood that the application of the present invention is not limited to the above for those of ordinary skills can
With improvement or transformation based on the above description, all these modifications and variations all should belong to the guarantor of appended claims of the present invention
Protect range.
Claims (10)
1. a kind of image motion artifact eliminating method based on artificial intelligence, which is characterized in that comprising steps of
Target image is handled to obtain the image with motion artifacts, and constructs image data base；
Neural network is constructed according to image data base, and carries out neural metwork training；
The motion artifacts of image to be processed are eliminated using trained neural network and export processed image.
2. the image motion artifact eliminating method according to claim 1 based on artificial intelligence, which is characterized in that described right
Target image handles to obtain the image with motion artifacts, and constructs image database steps and specifically include:
Two dimension is carried out to target image or three-dimensional Fourier transform forms Fourier space；
After increasing phase offset in the Fourier space of each angle, carries out inverse-Fourier transform and obtain with motion artifacts
Image；
The corresponding target image composition figure of image with motion artifacts is opposite, and form image data base.
3. the image motion artifact eliminating method according to claim 1 based on artificial intelligence, which is characterized in that described
Neural network is constructed according to image data base, and carries out neural metwork training step and specifically includes:
The neural network based on convolution is constructed according to image data base；
Using gradient descent method, according to the parameter of objective function optimization neural network, and complete to train.
4. the image motion artifact eliminating method according to claim 3 based on artificial intelligence, which is characterized in that the mesh
Scalar functions are as follows:
Wherein, Img is the image crossed by Processing with Neural Network, ImgtargFor target image, k is positive integer, ImgkFor kth
The image crossed by Processing with Neural Network, Imgtarg,kFor kth target image, Σ is sum operation, and Loss () is target letter
Number.
5. the image motion artifact eliminating method according to claim 3 based on artificial intelligence, which is characterized in that described
In neural network based on convolution, output background and various artifact regions when the image that output is crossed by Processing with Neural Network
Label.
6. a kind of image motion artifact based on artificial intelligence eliminates system characterized by comprising processor, and with institute
The memory of processor connection is stated,
The memory is stored with the image motion artifact based on artificial intelligence and eliminates program, the image based on artificial intelligence
Motion artifacts are eliminated when program is executed by the processor and are performed the steps of
Target image is handled to obtain the image with motion artifacts, and constructs image data base；
Neural network is constructed according to image data base, and carries out neural metwork training；
The motion artifacts of image to be processed are eliminated using trained neural network and export processed image.
7. the image motion artifact according to claim 6 based on artificial intelligence eliminates system, which is characterized in that the base
When the image motion artifact of artificial intelligence is eliminated program and executed by the processor, also perform the steps of
Two dimension is carried out to target image or three-dimensional Fourier transform forms Fourier space；
After increasing phase offset in the Fourier space of each angle, carries out inverse-Fourier transform and obtain with motion artifacts
Image；
The corresponding target image composition figure of image with motion artifacts is opposite, and form image data base.
8. the image motion artifact according to claim 6 based on artificial intelligence eliminates system, which is characterized in that the base
When the image motion artifact of artificial intelligence is eliminated program and executed by the processor, also perform the steps of
The neural network based on convolution is constructed according to image data base；
Using gradient descent method, according to the parameter of objective function optimization neural network, and complete to train.
9. the image motion artifact according to claim 8 based on artificial intelligence eliminates system, which is characterized in that the mesh
Scalar functions are as follows:
Wherein, Img is the image crossed by Processing with Neural Network, ImgtargFor target image, k is positive integer, ImgkFor kth
The image crossed by Processing with Neural Network, Imgtarg,kFor kth target image, Σ is sum operation, and Loss () is target letter
Number.
10. the image motion artifact according to claim 8 based on artificial intelligence eliminates system, which is characterized in that in institute
It states in the neural network based on convolution, output background and various artifact regions in the image that output is crossed by Processing with Neural Network
Label.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910325563.0A CN110060315B (en) | 2019-04-22 | 2019-04-22 | Image motion artifact eliminating method and system based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910325563.0A CN110060315B (en) | 2019-04-22 | 2019-04-22 | Image motion artifact eliminating method and system based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110060315A true CN110060315A (en) | 2019-07-26 |
| CN110060315B CN110060315B (en) | 2023-07-21 |
Family
ID=67320059
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910325563.0A Active CN110060315B (en) | 2019-04-22 | 2019-04-22 | Image motion artifact eliminating method and system based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110060315B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111223066A (en) * | 2020-01-17 | 2020-06-02 | 上海联影医疗科技有限公司 | Motion artifact correction method, motion artifact correction device, computer equipment and readable storage medium |
| CN111583354A (en) * | 2020-05-08 | 2020-08-25 | 上海联影医疗科技有限公司 | Training method for medical image processing unit and medical image motion estimation method |
| WO2021068643A1 (en) * | 2019-10-08 | 2021-04-15 | 南京安科医疗科技有限公司 | Composite energy spectrum ct imaging method |
| CN113052934A (en) * | 2021-03-16 | 2021-06-29 | 南开大学 | Nuclear magnetic resonance image motion artifact correction based on convolutional neural network |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB9614407D0 (en) * | 1996-07-09 | 1996-09-04 | Secr Defence | Method for imaging artefact reduction |
| CN104573746A (en) * | 2015-01-21 | 2015-04-29 | 中国计量学院 | Fruit fly type identification method based on magnetic resonance imaging |
| CN106056645A (en) * | 2016-05-25 | 2016-10-26 | 天津商业大学 | CT image translational motion artifact correction method based on frequency domain analysis |
| WO2017206048A1 (en) * | 2016-05-31 | 2017-12-07 | Shanghai United Imaging Healthcare Co., Ltd. | System and method for removing gibbs artifact in medical imaging system |
| CN107945132A (en) * | 2017-11-29 | 2018-04-20 | 深圳安科高技术股份有限公司 | A kind of artifact correction method and device of the CT images based on neutral net |
| US20190073804A1 (en) * | 2017-09-05 | 2019-03-07 | Siemens Healthcare Gmbh | Method for automatically recognizing artifacts in computed-tomography image data |
| US20190104940A1 (en) * | 2017-10-06 | 2019-04-11 | Toshiba Medical Systems Corporation | Apparatus and method for medical image reconstruction using deep learning for computed tomography (ct) image noise and artifacts reduction |
-
2019
- 2019-04-22 CN CN201910325563.0A patent/CN110060315B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB9614407D0 (en) * | 1996-07-09 | 1996-09-04 | Secr Defence | Method for imaging artefact reduction |
| CN104573746A (en) * | 2015-01-21 | 2015-04-29 | 中国计量学院 | Fruit fly type identification method based on magnetic resonance imaging |
| CN106056645A (en) * | 2016-05-25 | 2016-10-26 | 天津商业大学 | CT image translational motion artifact correction method based on frequency domain analysis |
| WO2017206048A1 (en) * | 2016-05-31 | 2017-12-07 | Shanghai United Imaging Healthcare Co., Ltd. | System and method for removing gibbs artifact in medical imaging system |
| US20190073804A1 (en) * | 2017-09-05 | 2019-03-07 | Siemens Healthcare Gmbh | Method for automatically recognizing artifacts in computed-tomography image data |
| US20190104940A1 (en) * | 2017-10-06 | 2019-04-11 | Toshiba Medical Systems Corporation | Apparatus and method for medical image reconstruction using deep learning for computed tomography (ct) image noise and artifacts reduction |
| CN107945132A (en) * | 2017-11-29 | 2018-04-20 | 深圳安科高技术股份有限公司 | A kind of artifact correction method and device of the CT images based on neutral net |
Non-Patent Citations (2)
| Title |
|---|
| 霍修坤等: ""一种心脏CT扫描图像重建算法研究"", 《计算机工程与应用》 * |
| 霍修坤等: ""一种心脏CT扫描图像重建算法研究"", 《计算机工程与应用》, 1 November 2005 (2005-11-01), pages 2 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021068643A1 (en) * | 2019-10-08 | 2021-04-15 | 南京安科医疗科技有限公司 | Composite energy spectrum ct imaging method |
| CN111223066A (en) * | 2020-01-17 | 2020-06-02 | 上海联影医疗科技有限公司 | Motion artifact correction method, motion artifact correction device, computer equipment and readable storage medium |
| CN111583354A (en) * | 2020-05-08 | 2020-08-25 | 上海联影医疗科技有限公司 | Training method for medical image processing unit and medical image motion estimation method |
| CN111583354B (en) * | 2020-05-08 | 2024-01-02 | 上海联影医疗科技股份有限公司 | Training method of medical image processing unit and medical image motion estimation method |
| CN113052934A (en) * | 2021-03-16 | 2021-06-29 | 南开大学 | Nuclear magnetic resonance image motion artifact correction based on convolutional neural network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110060315B (en) | 2023-07-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10859657B2 (en) | MRI reconstruction using deep learning, generative adversarial network and acquisition signal model | |
| CN110060315A (en) | A kind of image motion artifact eliminating method and system based on artificial intelligence | |
| Pezzotti et al. | An adaptive intelligence algorithm for undersampled knee MRI reconstruction | |
| CN107610194B (en) | Magnetic resonance image super-resolution reconstruction method based on multi-scale fusion CNN | |
| CN108537794B (en) | Medical image data processing method, apparatus and computer readable storage medium | |
| CN104933683B (en) | A kind of non-convex low-rank method for reconstructing for magnetic resonance fast imaging | |
| CN108416821B (en) | A kind of CT Image Super-resolution Reconstruction method of deep neural network | |
| CN109360152A (en) | 3 d medical images super resolution ratio reconstruction method based on dense convolutional neural networks | |
| Ye et al. | Deep residual learning for model-based iterative ct reconstruction using plug-and-play framework | |
| CN105022934B (en) | A kind of artificial immunity method for being used to build brain effective connectivity network from fMRI data | |
| Bhagat et al. | Data augmentation using generative adversarial networks for pneumonia classification in chest xrays | |
| CN110728729A (en) | Unsupervised CT projection domain data recovery method based on attention mechanism | |
| CN110298447B (en) | Method for processing parameters of machine learning method and reconstruction method | |
| CN107958471A (en) | CT imaging methods, device, CT equipment and storage medium based on lack sampling data | |
| CN112598649B (en) | 2D/3D spine CT non-rigid registration method based on generation of countermeasure network | |
| CN103218795A (en) | Partial K space sequence image reconstruction method based on self-adapted double-dictionary learning | |
| CN112348936A (en) | Low-dose cone-beam CT image reconstruction method based on deep learning | |
| CN114266939B (en) | Brain extraction method based on ResTLU-Net model | |
| CN115063309A (en) | Motion artifact simulator and corresponding method | |
| Chan et al. | An attention-based deep convolutional neural network for ultra-sparse-view CT reconstruction | |
| CN109389567A (en) | A kind of sparseness filtering method of fast optical imaging data | |
| CN106023124A (en) | Magnetic resonance imaging (MRI) sparse reconstruction image method based on prior image constraints | |
| CN112213673A (en) | Dynamic magnetic resonance imaging method, device, reconstruction computer and magnetic resonance system | |
| Song et al. | Super-resolution PET using a very deep convolutional neural network | |
| CN114926559A (en) | PET reconstruction method based on dictionary learning thought attenuation-free correction |
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 |