CN104597419A - Method for correcting motion artifacts in combination of navigation echoes and compressed sensing - Google Patents
Method for correcting motion artifacts in combination of navigation echoes and compressed sensing Download PDFInfo
- Publication number
- CN104597419A CN104597419A CN201510003289.7A CN201510003289A CN104597419A CN 104597419 A CN104597419 A CN 104597419A CN 201510003289 A CN201510003289 A CN 201510003289A CN 104597419 A CN104597419 A CN 104597419A
- Authority
- CN
- China
- Prior art keywords
- echo
- data
- image
- compressed sensing
- pseudorandom
- 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
Landscapes
- Magnetic Resonance Imaging Apparatus (AREA)
Abstract
The invention discloses a method for correcting motion artifacts in combination of navigation echoes and compressed sensing; the method comprises the steps: realizing a gradient echo sequence with navigation echoes and specially designing the phase encoding gradient applying order of the sequence; scanning a tested object by the gradient echo sequence to obtain original data; obtaining pseudo-random under-sampled k spatial data uninfluenced by motions based on the navigation echo information of the original data; performing image reconstruction to the pseudo-random under-sampled k spatial data by using the compressed sensing technology, and obtaining a reconstructed image; and performing multi-channel data combination to the reconstructed image in a mean square root manner and obtaining the corrected image.
Description
Technical field
The invention belongs to the technical field of magnetic resonance imaging, be specifically related to the motion artifacts antidote of a kind of navigation echo and compressed sensing.
Background technology
Motion is modal problem in magnetic resonance imaging, and the artifact caused by moving can reduce the quality of magnetic resonance image (MRI), has a strong impact on clinician and makes correct diagnosis.Navigator echo technology is a kind of technology being widely used in monitoring imaging object motion and artifact rectification.Within the scope of certain amplitude, navigator echo can the motion of effective tracking objects, as respiratory movement, by a small margin head movement etc., and can utilize and monitors the movable information obtained and correct k-space data, thus reduce motion artifacts to the impact of picture quality.But navigator echo technology cannot correct the image artifacts that Large Amplitude Motion and non-rigid motion cause, the k-space data that the motion moment collects can only be given up.
Compressed sensing (Compressed Sensing, CS), by non-linear reconstruction algorithm, under the condition much smaller than Nyquist sampling rate, is carried out reconstruction to the data of lack sampling and is recovered original signal.Image reconstruction demand fulfillment two essential condition are carried out in applied compression perception: the image that (1) will rebuild can transform to a kind of sparse form; (2) k-space data is necessary for random lack sampling, and it is after Fourier transform, and the artifact in image is noncoherent as noise.Magnetic resonance image (MRI) can meet this two conditions well, therefore compressed sensing technology can be applied to magnetic resonance imaging arts.In recent years, compressed sensing is widely used in magnetic resonance imaging arts, rebuilds the data of random lack sampling, to shorten imaging time.Barral group proposes to carry out motion artifacts rectification in conjunction with compressed sensing and navigator echo technology, but they just correct for the situation of getting back to origin-location after motion, this has significant limitation in the application of reality, because tested object is difficult to get back to initial position after movement.
Summary of the invention
The present invention is directed to above-mentioned navigator echo and cannot correct the weak points such as the artifact that Large Amplitude Motion or non-rigid motion cause, the motion artifacts antidote of a kind of navigation echo and compressed sensing is proposed, the method effectively can correct the artifact that Large Amplitude Motion and non-rigid motion cause, and testedly must get back to original position after movement without the need to requiring.For realizing above object, the present invention proposes following technical scheme:
The present invention proposes the motion artifacts antidote of a kind of navigation echo and compressed sensing, said method comprising the steps of:
Step a: realize the gradin-echo having navigator echo, specifically designs the phase encoding gradient applying order of described sequence;
Step b: with described gradin-echo, scanning is carried out to tested object and obtain raw data;
Step c: obtain not by the pseudorandom lack sampling k-space data of motion effects based on the raw data obtained in step b;
Steps d: carry out image reconstruction to described pseudorandom lack sampling k-space data by compressed sensing technology, obtains and rebuilds image; And
Step e: carry out multi-channel data combination to described reconstruction image by root mean square mode, obtains the image after correcting.
The present invention propose described navigation echo and compressed sensing motion artifacts antidote in, in step a using the echo of second in described gradin-echo as navigator echo, for the movable information of detecting object.
In the motion artifacts antidote of the described navigation echo that the present invention proposes and compressed sensing, in step a, describedly specific design is carried out to phase encoding gradient applying order be k-space data acquisition order is specifically designed, comprise: the collection adopting twice k-space immediate vicinity data, namely gather in scanning incipient stage and ending phase respectively, other k-space data carries out random acquisition in the phase encode direction.
In the motion artifacts antidote of the described navigation echo that the present invention proposes and compressed sensing, obtain pseudorandom lack sampling k-space data in step c and comprise further:
Step c1: described raw data is divided into navigator echo data and image echo data;
Step c2: with described navigator echo data acquisition movable information, combines described movable information and described phase encoding gradient applying order and obtains pseudorandom lack sampling template; And
Step c3: obtained not by the pseudorandom lack sampling k-space data of motion effects by described pseudorandom lack sampling template and described image echo data.
In the motion artifacts antidote of the described navigation echo that the present invention proposes and compressed sensing, in step c, obtain movable information to select the part not being subject to the data of motion effects more as described pseudorandom lack sampling k-space data with navigator echo.
In the motion artifacts antidote of the described navigation echo that the present invention proposes and compressed sensing, be multiplied by described image echo data by described pseudorandom lack sampling template in step c3 and obtain described pseudorandom lack sampling k-space data.
In the motion artifacts antidote of the described navigation echo that the present invention proposes and compressed sensing, steps d also comprises the steps:
Steps d 1: this k-space data, after inverse-Fourier transform, obtains undersampled image m; And
Steps d 2: compressed sensing reconstruction is carried out to undersampled image m.
The inventive method is compared with the pseudo-image method of traditional navigator echo corrective exercise, does not need to consider how object moves and whether meet rigid motion.Owing to being removed by the data of motion effects, therefore the rectification effect of artifact has nothing to do with the amplitude of motion and form, and the problem that Large Amplitude Motion or non-rigid motion cannot be corrected obtains good solution.The present invention has carried out specific design to phase encoding gradient applying order, there is the collection of the k-space centre data of twice, testedly must get back to origin-location after movement without the need to requiring like this, can select not rebuild by a part of data that the data volume of motion effects is more.Meanwhile, this method is suitable for other sequence, and the motion artifacts for magnetic resonance imaging is corrected and provided new approaches.
Accompanying drawing explanation
Fig. 1 is the 2-dimensional gradient echo pulse sequence timing diagram being integrated with navigator echo according to realizing in embodiments of the invention one.
Fig. 2 applies precedence diagram according to the phase encoding gradient of two-dimensional sequence in embodiments of the invention one.
Fig. 3 is the process flow diagram according to motion artifacts antidote in embodiments of the invention.
Fig. 4 is lack sampling template in the embodiment of the present invention one in two-dimensional imaging and corresponding lack sampling k-space data, and Fig. 4 (a) illustrates two-dimentional lack sampling template, and Fig. 4 (b) illustrates two-dimentional lack sampling k-space.
Fig. 5 is the whole iterative process figure during compressed sensing is rebuild.
Fig. 6 is the two-dimentional cranium mind map before and after the rectification that obtains according to the motion artifacts antidote of the embodiment of the present invention one, Fig. 6 (a) illustrates that scanning head in early stage moves, Fig. 6 (b) illustrates that scanning later stage head moves, and Fig. 6 (c) illustrates that random once head moves.
Fig. 7 is the 3-dimensional gradient echo pulse sequence timing diagram being integrated with navigator echo according to realizing in embodiments of the invention two.
Fig. 8 applies precedence diagram according to the phase encoding gradient of three-dimensional series in embodiments of the invention two.
Fig. 9 is lack sampling template in the embodiment of the present invention two in three-dimensional imaging and corresponding lack sampling k-space data, and Fig. 9 (a) illustrates three-dimensional lack sampling template, and Fig. 9 (b) illustrates three-dimensional lack sampling k-space.
Figure 10 is the 3d knee joint figure before and after the rectification that obtains according to the motion artifacts antidote of the embodiment of the present invention two, Figure 10 (a) illustrates scanning motion in early stage, Figure 10 (b) scans anaphase movement, and Figure 10 (c) once moves at random.
Embodiment
In conjunction with following specific embodiments and the drawings, the invention will be described in further detail.Implement process of the present invention, condition, experimental technique etc., except the following content mentioned specially, be universal knowledege and the common practise of this area, the present invention is not particularly limited content.
The inventive method has corrected the artifact that the Large Amplitude Motion of any time or non-rigid motion cause preferably, without the need to Resurvey.Two embodiments step-by-step instructions the inventive method sequences Design respectively, obtains pseudorandom lack sampling k-space data, carries out the specific operation process of image reconstruction by compressed sensing technology below.
Embodiment one:
In the present embodiment, the magnetic resonance imaging data of collection is the cranium brain magnetic resonance imaging data of two dimension.Gradient echoes sequence realizes on the sequence integrating and developing platform IDEA of Siemens Company, and magnetic resonance imaging data is captured on Siemens MAGNETOM Trioa Tim 3T magnetic resonance imaging system and completes.
The concrete imaging parameters of 2-dimensional gradient echo is: repetition time (TR)=250ms, the echo time (TE) of image echo is 4.8ms, the TE of navigator echo is 11ms, flip angle (FA)=70 °, observe wild (FOV)=250mm × 250mm (frequency coding × phase encoding), thickness=5mm, bandwidth (BW)=330Hz/pixel, scan matrix (matrix)=256 × 286 (frequency coding × phase encoding).
Below in conjunction with accompanying drawing 1-6, the motion artifacts antidote according to the embodiment of the present invention one is specifically described.Fig. 3 is the process flow diagram according to motion artifacts antidote in embodiments of the invention, is described below to the step a-e of antidote.
Step a: realize the gradin-echo having navigator echo, specifically designs phase encoding gradient applying order.
With reference to the 2-dimensional gradient echo pulse sequence timing diagram being integrated with navigator echo that figure 1, Fig. 1 is according to realizing in embodiments of the invention one.In Fig. 1: RF is low-angle radio frequency; GS is slice selective gradient; GP is phase encoding gradient; GR is readout direction gradient, adc data collection.First echo collected is image echo, and before collection second echo, the effect of phase encoding gradient is offset by corresponding oppositely gradient, and therefore, second echo is the one dimension navigator echo not having phase information, can be used for the motion of detecting object.
In Fig. 1, in this 2-dimensional gradient echo pulse train, phase encoding is pseudorandom.In scanning incipient stage and ending phase, repeated acquisition twice k-space immediate vicinity data; And the applying order random alignment of the phase encoding gradient at other non-k-space centers, realize non-k-space data random acquisition in the phase encode direction.
Step b: have the tested cranium brain of the 2-dimensional gradient echo sequence pair of navigator echo to carry out scanning to obtain raw data with above-mentioned.In order to verify validity of the present invention, the present embodiment carries out three scanning to tested cranium brain, require that tested cranium brain carries out once significantly head movement in each scanning, head to move when time point is respectively scanning when starting, the end of scan and random time in the middle of single pass.
Step c: obtain not by the pseudorandom lack sampling k-space data of motion effects based on the raw data obtained in step b.Specifically comprise: step c1: by the raw data obtained in step b, be divided into navigator echo data and image echo data two parts; Step c2: with the movable information of object during described navigator echo data acquisition image data, combines described movable information and described phase encoding gradient applying order and obtains pseudorandom lack sampling template; Step c3: obtained not by the pseudorandom lack sampling k-space data of motion effects by the raw k-space data of described pseudorandom lack sampling template and described image echo.
It is the two-dimensional pseudo-random lack sampling template obtained in the present embodiment shown in Fig. 4 a.Kx direction is frequency coding direction, for entirely adopting; Ky direction is phase-encoding direction, for owing to adopt.Wherein black represents 0, corresponding to the data of k-space by motion effects, will be rejected during reconstruction; White represents 1, corresponding to k-space not by the data of motion effects, will be used for image reconstruction.As shown in Figure 4 b, the raw k-space data of image echo data are multiplied by lack sampling template and obtain pseudorandom lack sampling k-space data, and k-space data now is not all subject to motion effects.
Steps d: carry out image reconstruction to described pseudorandom lack sampling k-space data by compressed sensing technology, obtains and rebuilds image.
This steps d specifically also comprises the steps d1 and steps d 2.
Steps d 1: this k-space data, after inverse-Fourier transform, obtains undersampled image m.Now the motion artifacts obtained in step b is transformed to the incoherent artifact that pseudorandom lack sampling causes, and this incoherent artifact problem solves by compressed sensing reconstruction.
Steps d 2: compressed sensing reconstruction is carried out to undersampled image m.
Fig. 5 is the process flow diagram of the whole iterative process during compressed sensing is rebuild.
Finally, the image after being rebuild by all passages adopts root mean square mode to carry out Multichannel combination (S37), the image after can finally being corrected.
Theoretical model such as the formula [1] of compressed sensing image reconstruction is expressed as follows:
minimize||Ψm||
1+λTV(m)[1]
Wherein, matrix m represents a width magnetic resonance image (MRI) of reconstruction; Operational character Ψ represents certain sparse transformation, image m can be transformed to a kind of rarefaction representation; Matrix Ψ m refers to the image of image m after Ψ converts; TV (Total Variation) represents the total variation of image, and also as a kind of sparse transformation, λ is weight;
represent and first two-dimensional Fourier transform is carried out to image, after obtaining k-space data, then lack sampling operation is carried out to k-space data; Y represents the lack sampling data of real scan; ∈ is a small quantity, represents the accuracy requirement of the difference of data reconstruction and true image data.
Below in conjunction with Fig. 5, iterative process is specifically described.First, undersampled image m obtains wavelet coefficient (S501) after wavelet transformation Ψ, by Nonlinear conjugate gradient algorithm, is optimized energy function, and energy function is made up of fidelity item, small echo item and TV item.Wavelet coefficient obtains new images (S502) through anti-wavelet transformation, new k-space data (S503) is obtained by inverse-Fourier transform, owe to adopt to these data by same lack sampling template, obtain new lack sampling k-space data (S504), pass through constraint condition with real k-space data y
carry out the judgement of data fidelity, obtain fidelity item (S505); New images is obtained after anti-wavelet transformation to wavelet coefficient and carries out image total variation (S506), and ask L1 norm to obtain TV item λ TV (m) (S507); Ask L1 norm can obtain small echo item to wavelet coefficient || Ψ m||
1(S508), then the result of S505, S507, S508 tri-steps is sued for peace, calculate new energy function En (S509).If energy function increases, namely condition En<En-1 (S510) cannot meet, then readjust wavelet coefficient (S501), computation energy function again.If energy function reduces, namely condition En<En-1 (S510) is met, then using the energy function of this calculating as new energy function, then rebuild number of times and add 1, upgrade wavelet coefficient (S511), until energy function variable quantity be less than certain threshold value or reach the iterations of setting after (S512) jump out circulation, iteration terminates.In this example, main reconstruction parameter is as follows: iterations=3*8, TV penalty term weight=0.002, L1 norm penalty term weight=0.005.Resulting in the figure after compressed sensing reconstruction, the incoherent artifact caused by pseudorandom lack sampling now obtains suppression.
Step e: carry out multi-channel data combination to described reconstruction image by root mean square mode, obtains the image after correcting.Particularly, scan in the raw data obtained in step b, two-dimentional cranium brain image has the data of four groups of passages, by steps d, above-mentioned image reconstruction is carried out to the data of each passage, data after reconstruction are plural number, ask quadratic sum to open root again real and imaginary part, obtain the modulus value of image; Open root again by asking quadratic sum to the modulus value of multiple passage and carry out combination of channels, the image after finally being corrected.
In the present embodiment, two-dimensional imaging result as Fig. 6 correct before shown in, motion cause image to produce serious artifact, G-WMI is unclear, does not meet medical diagnosis requirement, needs to carry out motion artifacts rectification.Correction result as Fig. 6 correct after shown in, G-WMI is clear, well shows the fine structure of cranium brain.Fig. 6 (a) illustrates that scanning head in early stage moves, and Fig. 6 (b) illustrates that scanning later stage head moves, and Fig. 6 (c) illustrates that random once head moves.As seen from Figure 6, the artifact that the motion of different time sections causes is obtained for suppression preferably or reduces, and the image after rectification all reaches medical diagnosis requirement.
Embodiment two:
What the embodiment of the present invention two and the difference of embodiment one were to gather is three-dimensional imaging data, is in addition identical on motion artifacts antidote.
The magnetic resonance imaging data that embodiment two gathers is three-dimensional MRI of knee data.Sequence realizes on the sequence integrating and developing platform IDEA of Siemens Company, and data acquisition completes on Siemens MAGNETOM Trio a Tim 3T magnetic resonance imaging system.
The concrete imaging parameters of 3-dimensional gradient echo is: TR=30ms, the TE of image echo is 4.8ms, the TE of navigator echo is 11ms, FA=15 °, FOV=256mm × 128mm (frequency coding × phase encoding), thickness=1mm, BW=390Hz/pixel, matrix=256 × 136 × 128 (frequency coding × phase encoding × layer direction phase encoding).
The step of the motion artifacts antidote according to the embodiment of the present invention two is illustrated below in conjunction with accompanying drawing 3,5-10.Fig. 3 is the process flow diagram according to motion artifacts antidote in embodiments of the invention, is described below to the step a-e of antidote.
Step a: realize the gradin-echo having navigator echo, specifically designs phase encoding gradient applying order.
In gradin-echo, second echo is as navigator echo, selects on layer direction and is offset by corresponding oppositely gradient with the effect of the phase encoding gradient on phase-encoding direction.This echo is one dimension navigator echo, can be used for the motion of detecting object, as shown in Figure 7.
In Fig. 7, in this 3-dimensional gradient echo pulse train, phase-encoding direction is pseudo-random permutation with the phase encoding selected on layer direction.In scanning incipient stage and ending phase, repeated acquisition twice k-space immediate vicinity data; And the applying order random alignment of the phase encoding gradient at other non-k-space center, realize non-k-space data random acquisition in the phase encode direction.Phase encoding gradient applying order as shown in Figure 8.
Step b: have the tested knee joint of 3-dimensional gradient echo sequence pair of navigator echo to carry out scanning to obtain raw data with above-mentioned.In order to verify validity of the present invention, this example carries out three scanning to tested, requires testedly in each scanning, to carry out once significantly motion of knee joint, random time when run duration point is respectively scanning when starting, the end of scan and in the middle of single pass.
Step c: obtain not by the pseudorandom lack sampling k-space data of motion effects based on the raw data obtained in step b.Specifically comprise: step c1: by the raw data obtained in step b, be divided into navigator echo data and image echo data two parts; Step c2: with the movable information of object during described navigator echo data acquisition image data, combines described movable information and described phase encoding gradient applying order and obtains pseudorandom lack sampling template; Step c3: obtained not by the pseudorandom lack sampling k-space data of motion effects by the raw k-space data of described pseudorandom lack sampling template and described image echo.
It is the three-dimensional pseudorandom lack sampling template obtained in the present embodiment shown in Fig. 9 a.Kx direction is frequency coding direction, for entirely adopting; Ky direction is phase-encoding direction, and kz direction, for selecting layer direction, is and owes to adopt. wherein black represents 0, needs the data be rejected corresponding to k-space; White represents 1, and represent the data being used for image reconstruction, sample template is the matrix of 128 × 128.At the Data processing of 3-D view, need first on frequency coding direction, to carry out one dimensional fourier transform, then to the party upwards the k-space of every layer use same sample template, raw k-space data obtain lack sampling k-space data under sample template effect.Fig. 9 b is depicted as the lack sampling k-space data in the present embodiment.
Steps d: carry out image reconstruction to described pseudorandom lack sampling k-space data by compressed sensing technology, obtains and rebuilds image.
This steps d specifically also comprises the steps d1 and steps d 2.
Steps d 1: this k-space data, after inverse-Fourier transform, obtains undersampled image m.Now the motion artifacts obtained in step b is transformed to the incoherent artifact that pseudorandom lack sampling causes, and this incoherent artifact problem solves by compressed sensing reconstruction.
Steps d 2: compressed sensing reconstruction is carried out to undersampled image m.
Figure 5 shows that the whole iterative process (this process is described in detail in embodiment one, does not repeat them here) that in the present embodiment, compressed sensing is rebuild.In this example, main reconstruction parameter is as follows: iterations=3*8, TV penalty term weight=0.002, L1 norm penalty term weight=0.005.
Resulting in the figure after compressed sensing reconstruction, the incoherent artifact caused by pseudorandom lack sampling now obtains suppression.
Step e: carry out multi-channel data combination to described reconstruction image by root mean square mode, obtains the image after correcting.Particularly, scan in the raw data obtained in step b, 3d knee joint image has the data of eight groups of passages, by steps d, above-mentioned image reconstruction is carried out to the data of each passage, data after reconstruction are plural number, ask quadratic sum to open root again real and imaginary part, obtain the modulus value of image; Open root again by asking quadratic sum to the modulus value of multiple passage and carry out combination of channels, the image after finally being corrected.
In the present embodiment, three-dimensional imaging result as Figure 10 correct before shown in, can see the artifact of obvious similar noise, especially on transversal section, the details of image is covered, musculature is coarse, institutional framework boundary is fuzzy, has had a strong impact on contrast and the resolution of image, has been easy to cover little focus, affect medical diagnosis, therefore need to carry out motion artifacts rectification.Correction result as Figure 10 correct after shown in, demarcate between institutional framework clear, musculature is level and smooth, and details is clearly demarcated.Figure 10 (a) illustrates scanning motion in early stage, and Figure 10 (b) scans anaphase movement, and Figure 10 (c) once moves at random.As seen from Figure 10, the artifact that the motion of different time sections causes is obtained for suppression preferably or reduces, and the image after rectification all reaches medical diagnosis requirement.
Protection content of the present invention is not limited to above embodiment.Under the spirit and scope not deviating from inventive concept, the change that those skilled in the art can expect and advantage are all included in the present invention, and are protection domain with appending claims.
Claims (7)
1. a motion artifacts antidote for navigation echo and compressed sensing, is characterized in that, said method comprising the steps of:
Step a: realize the gradin-echo having navigator echo, specifically designs the phase encoding gradient applying order of described sequence;
Step b: with described gradin-echo, scanning is carried out to tested object and obtain raw data;
Step c: obtain not by the pseudorandom lack sampling k-space data of motion effects based on the raw data obtained in step b;
Steps d: carry out image reconstruction to described pseudorandom lack sampling k-space data by compressed sensing technology, obtains and rebuilds image; And
Step e: carry out multi-channel data combination to described reconstruction image by root mean square mode, obtains the image after correcting.
2. the motion artifacts antidote of navigation echo as claimed in claim 1 and compressed sensing, is characterized in that, in step a using the echo of second in described gradin-echo as navigator echo, for the movable information of detecting object.
3. the motion artifacts antidote of navigation echo as claimed in claim 1 and compressed sensing, it is characterized in that, in step a, describedly specific design is carried out to phase encoding gradient applying order be k-space data acquisition order is specifically designed, comprise: the collection adopting twice k-space immediate vicinity data, namely gather in scanning incipient stage and ending phase respectively, other k-space data carries out random acquisition in the phase encode direction.
4. the motion artifacts antidote of navigation echo as claimed in claim 1 and compressed sensing, is characterized in that, obtain pseudorandom lack sampling k-space data and comprise further in step c:
Step c1: described raw data is divided into navigator echo data and image echo data;
Step c2: with described navigator echo data acquisition movable information, combines described movable information and described phase encoding gradient applying order and obtains pseudorandom lack sampling template; And
Step c3: obtained not by the pseudorandom lack sampling k-space data of motion effects by described pseudorandom lack sampling template and described image echo data.
5. the motion artifacts antidote of navigation echo as claimed in claim 4 and compressed sensing, it is characterized in that, in step c, obtain movable information to select the part not being subject to the data of motion effects more as described pseudorandom lack sampling k-space data with navigator echo.
6. the motion artifacts antidote of navigation echo as claimed in claim 4 and compressed sensing, is characterized in that, be multiplied by described image echo data obtain described pseudorandom lack sampling k-space data in step c3 by described pseudorandom lack sampling template.
7. the motion artifacts antidote of navigation echo as claimed in claim 1 and compressed sensing, it is characterized in that, steps d also comprises the steps:
Steps d 1: this k-space data, after inverse-Fourier transform, obtains undersampled image m; And
Steps d 2: compressed sensing reconstruction is carried out to undersampled image m.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510003289.7A CN104597419A (en) | 2015-01-04 | 2015-01-04 | Method for correcting motion artifacts in combination of navigation echoes and compressed sensing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510003289.7A CN104597419A (en) | 2015-01-04 | 2015-01-04 | Method for correcting motion artifacts in combination of navigation echoes and compressed sensing |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104597419A true CN104597419A (en) | 2015-05-06 |
Family
ID=53123325
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510003289.7A Pending CN104597419A (en) | 2015-01-04 | 2015-01-04 | Method for correcting motion artifacts in combination of navigation echoes and compressed sensing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104597419A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106338703A (en) * | 2016-09-30 | 2017-01-18 | 中国科学院武汉物理与数学研究所 | Radio frequency pulse multimode weighting-based high-resolution fast magnetic resonance imaging method |
| CN108535674A (en) * | 2018-02-06 | 2018-09-14 | 苏州朗润医疗系统有限公司 | Mitigate the method for artifact for the multiple averaging of fast spin echo |
| CN110652296A (en) * | 2019-09-16 | 2020-01-07 | 华东师范大学 | Method for removing magnetic resonance head image motion artifact |
| CN110794352A (en) * | 2018-08-02 | 2020-02-14 | 上海康达卡勒幅医疗科技有限公司 | Method for inhibiting magnetic resonance image artifact |
| CN110992435A (en) * | 2019-11-06 | 2020-04-10 | 上海东软医疗科技有限公司 | Image reconstruction method and device, and imaging data processing method and device |
| CN112204389A (en) * | 2018-03-29 | 2021-01-08 | 筑波科技株式会社 | Image processing method for ultrasonic transmission image |
| CN117876279A (en) * | 2024-03-11 | 2024-04-12 | 浙江荷湖科技有限公司 | Method and system for removing motion artifact based on scanned light field sequence image |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101509964A (en) * | 2009-03-06 | 2009-08-19 | 华东师范大学 | Method for correcting instability of main magnetic field in magnetic resonance imaging system |
| CN102540115A (en) * | 2011-12-12 | 2012-07-04 | 中国科学院深圳先进技术研究院 | Magnetic resonance imaging method and system |
| CN103809141A (en) * | 2012-11-02 | 2014-05-21 | 弗赖堡大学综合诊所 | Method for magnetic resonance imaging |
| CN104020430A (en) * | 2014-04-15 | 2014-09-03 | 清华大学 | Correction method and system for movement artifacts of magnetic resonance imaging |
-
2015
- 2015-01-04 CN CN201510003289.7A patent/CN104597419A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101509964A (en) * | 2009-03-06 | 2009-08-19 | 华东师范大学 | Method for correcting instability of main magnetic field in magnetic resonance imaging system |
| CN102540115A (en) * | 2011-12-12 | 2012-07-04 | 中国科学院深圳先进技术研究院 | Magnetic resonance imaging method and system |
| CN103809141A (en) * | 2012-11-02 | 2014-05-21 | 弗赖堡大学综合诊所 | Method for magnetic resonance imaging |
| CN104020430A (en) * | 2014-04-15 | 2014-09-03 | 清华大学 | Correction method and system for movement artifacts of magnetic resonance imaging |
Non-Patent Citations (1)
| Title |
|---|
| 董芳: "结合导航回波技术进行伪影矫正研究", 《硕士学位论文集2014》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106338703B (en) * | 2016-09-30 | 2018-12-25 | 中国科学院武汉物理与数学研究所 | A kind of high definition rapid magnetic resonance imaging method based on the weighting of radio-frequency pulse multimode |
| CN106338703A (en) * | 2016-09-30 | 2017-01-18 | 中国科学院武汉物理与数学研究所 | Radio frequency pulse multimode weighting-based high-resolution fast magnetic resonance imaging method |
| CN108535674B (en) * | 2018-02-06 | 2020-11-20 | 苏州朗润医疗系统有限公司 | Method for reducing artifacts by multiple averaging of fast spin echoes |
| CN108535674A (en) * | 2018-02-06 | 2018-09-14 | 苏州朗润医疗系统有限公司 | Mitigate the method for artifact for the multiple averaging of fast spin echo |
| CN112204389B (en) * | 2018-03-29 | 2023-03-28 | 筑波科技株式会社 | Image processing method for ultrasonic transmission image |
| CN112204389A (en) * | 2018-03-29 | 2021-01-08 | 筑波科技株式会社 | Image processing method for ultrasonic transmission image |
| CN110794352B (en) * | 2018-08-02 | 2023-05-26 | 上海康达卡勒幅医疗科技有限公司 | Method for inhibiting magnetic resonance image artifact |
| CN110794352A (en) * | 2018-08-02 | 2020-02-14 | 上海康达卡勒幅医疗科技有限公司 | Method for inhibiting magnetic resonance image artifact |
| CN110652296A (en) * | 2019-09-16 | 2020-01-07 | 华东师范大学 | Method for removing magnetic resonance head image motion artifact |
| CN110992435A (en) * | 2019-11-06 | 2020-04-10 | 上海东软医疗科技有限公司 | Image reconstruction method and device, and imaging data processing method and device |
| CN110992435B (en) * | 2019-11-06 | 2023-10-20 | 上海东软医疗科技有限公司 | Image reconstruction method and device, imaging data processing method and device |
| CN117876279A (en) * | 2024-03-11 | 2024-04-12 | 浙江荷湖科技有限公司 | Method and system for removing motion artifact based on scanned light field sequence image |
| CN117876279B (en) * | 2024-03-11 | 2024-05-28 | 浙江荷湖科技有限公司 | Method and system for removing motion artifact based on scanned light field sequence image |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104597419A (en) | Method for correcting motion artifacts in combination of navigation echoes and compressed sensing | |
| CN104068859B (en) | For determining method and the magnetic resonance equipment of multiple magnetic resonance image (MRI) | |
| Cheng et al. | Highly scalable image reconstruction using deep neural networks with bandpass filtering | |
| CN101975935B (en) | Partial echo compressed sensing-based quick magnetic resonance imaging method | |
| CN106997034B (en) | Based on the magnetic resonance diffusion imaging method rebuild using Gauss model as example integration | |
| Anderson III et al. | Adaptive retrospective correction of motion artifacts in cranial MRI with multicoil three‐dimensional radial acquisitions | |
| CN102540116A (en) | Magnetic resonance imaging method and system | |
| CN106461746A (en) | Multi-echo mri using repeated sampling of k-lines with different sequential order per repetition | |
| CN109115820B (en) | Magnetic resonance water-fat separation and quantification method and device based on plane echo imaging | |
| CN105074491A (en) | Dynamic MRI with image reconstruction using compressed sensing | |
| Gao et al. | PCLR: phase‐constrained low‐rank model for compressive diffusion‐weighted MRI | |
| CN105143906A (en) | Metal resistant MR imaging | |
| CN110133556A (en) | A kind of magnetic resonance image processing method, device, equipment and storage medium | |
| CN102204818A (en) | Magnetic resonance imaging method | |
| CN105467339A (en) | Quick multilayer magnetic resonance imaging method and device | |
| CN109212443A (en) | The equal voxels magnetic resonance diffusion imaging method and device excited simultaneously based on more plates | |
| CN105931242B (en) | Dynamic nuclear magnetic resonance (DNMR) image rebuilding method based on dictionary learning and time gradient | |
| CN107205686A (en) | Remove in the method for the ghost image artifact of magnetic resonance imaging and for this magnetic resonance device | |
| US9208587B2 (en) | Systems and methods for compressed sensing for multi-shell magnetic resonance imaging | |
| CN102590773A (en) | Magnetic resonance imaging method and system | |
| CN107110945A (en) | For the MRI agreements split using the image gathered under two kinds of different magnetic field intensity to image detail | |
| CN105678822A (en) | Three-regular magnetic resonance image reconstruction method based on Split Bregman iteration | |
| CN102540115A (en) | Magnetic resonance imaging method and system | |
| CN104181484A (en) | Magnetic resonance image reconstruction method | |
| Hou et al. | PNCS: Pixel-level non-local method based compressed sensing undersampled MRI image reconstruction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20150506 |