CN107092805A - magnetic resonance parallel imaging device - Google Patents
magnetic resonance parallel imaging device Download PDFInfo
- Publication number
- CN107092805A CN107092805A CN201710368811.0A CN201710368811A CN107092805A CN 107092805 A CN107092805 A CN 107092805A CN 201710368811 A CN201710368811 A CN 201710368811A CN 107092805 A CN107092805 A CN 107092805A
- Authority
- CN
- China
- Prior art keywords
- data
- obk
- vector
- gathered
- refinement
- 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
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R33/00—Arrangements or instruments for measuring magnetic variables
- G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
- G01R33/44—Arrangements or instruments for measuring magnetic variables involving magnetic resonance using nuclear magnetic resonance [NMR]
- G01R33/48—NMR imaging systems
- G01R33/54—Signal processing systems, e.g. using pulse sequences ; Generation or control of pulse sequences; Operator console
- G01R33/56—Image enhancement or correction, e.g. subtraction or averaging techniques, e.g. improvement of signal-to-noise ratio and resolution
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
- G16H40/60—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices
- G16H40/63—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices for local operation
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Primary Health Care (AREA)
- Epidemiology (AREA)
- Medical Informatics (AREA)
- General Business, Economics & Management (AREA)
- Public Health (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Signal Processing (AREA)
- Business, Economics & Management (AREA)
- High Energy & Nuclear Physics (AREA)
- Condensed Matter Physics & Semiconductors (AREA)
- General Physics & Mathematics (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
Abstract
The invention provides a kind of magnetic resonance parallel imaging device, including：Collecting unit, for gathering k-space data, the k-space data includes gathered data and calibration data；Computing unit, obtains refining data, and obtain coil merging coefficient by the refinement data and calibration data calculating for carrying out the gathered data data mining；Shim, for merging coefficient and the gathered data according to the coil, fills up and obtains complete k-space data；Imaging unit, image is obtained for the complete k-space data to be transformed into image area.Pass through technical scheme provided by the present invention, the convolution kernel chosen in process of reconstruction is gathered for magnetic resonance parallel, carry out data and bring up again refining, so that the data volume after refining reduces, signal characteristic is strengthened, using the convolution kernel after refinement, data filling is carried out, new convolution kernel is obtained and is optimized than original.
Description
The application be based on be submitted on January 9th, 2014 Patent Office of the People's Republic of China, Application No. 201410010303.1, invention
Point of the Chinese patent application of entitled " magnetic resonance coil merges coefficient calculation method, MR imaging method and its device "
Case.
Technical field
The present invention relates to magnetic resonance imaging arts, more particularly to a kind of magnetic resonance parallel imaging device.
Background technology
In mr imaging technique, the speed of imaging is to weigh a critically important standard of imaging method.It is constrained to picture
The critically important factor of speed is data acquisition, and k-space filling.General data acquisition modes will adopt full k-space data, so
After could rebuild obtaining image.Magnetic resonance parallel gathers reconstruction technique, is that the mode merged is recombinated using coil, to owing to adopt
The data of sample are filled up, and are rebuild using the k-space data for filling up complete.Profit in such a way, can according to demand,
Only gather a part of k-space data, it is not necessary to adopt completely whole k-space.Therefore such method can greatly speed up the speed of imaging.
One of the more commonly used method for parallel reconstruction is GRAPPA.Traditional GRAPPA algorithm is as shown in figure 1, black real point
It is represented as the k-space data of actual acquisition；White null point is the data that lack sampling needs to fill up；Grey real point represents calculating
Coil merging parameter, and the appropriate calibration data adopted entirely.GRAPPA algorithm thinks that the hollow dots of any one in figure can be expressed as
The linear superposition of surrounding black real point, is merged equivalent to the data to multiple coils.And coil merges coefficient nij(i-th
Individual coil, such as j-th of position, Fig. 1) it can be fitted Grey Point to determine by the real point of black.Coil merges after coefficient determination,
Other white hollow points can be calculated to fill up and obtained according to the merging coefficient and black real point tried to achieve, and obtain complete so as to rebuild
K-space data.
Coil merges coefficient and can be described as convolution kernel again, and in conventional methods where, the calculated direction of convolution kernel only adds phase
Encode direction, and channel direction.For effect of optimization, in recent years, many methods introduce other directions such as frequency coding side
To, and time orientation etc. in k-t.The introducing of these dimensions, making the information of convolution kernel increases, but with excessive introducing,
Redundancy occurs in information, and due to the influence of noise so that the coefficient of calculating nevertheless suffers from influence；And convolution kernel is excessive, make
The coefficient that must be fitted increases, and also brings the unstability of calculating.And convolution kernel is too small, then data message is not enough, calculating process
It is not accurate enough.
The content of the invention
The problem to be solved in the present invention is to provide a kind of magnetic resonance parallel imaging device, solves during convolution kernel selection
Due to convolution kernel choose it is excessive cause information content to increase, information redundancy, and noise is excessive, fitting coefficient increases, and calculates not
Stable influence, and when convolution kernel chooses too small, data deficiencies is calculated, the problem of as a result inaccurate.
The invention provides a kind of magnetic resonance parallel imaging device, including：
Collecting unit, suitable for collection k-space data, the k-space data includes gathered data and calibration data；
Computing unit, obtains refining data, the gathered data is being carried suitable for carrying out the gathered data data mining
It is n to refine the dimension on direction, and dimension of the refinement data on the refinement direction is t, and described t, n are positive integer and t<
N, and coil merging coefficient is obtained by the refinement data and calibration data calculating；
Shim, suitable for merging coefficient and the gathered data according to the coil, fills up and obtains complete k-space number
According to；
Imaging unit, image is obtained suitable for the complete k-space data is transformed into image area；
The computing unit is suitable to：
By vector based on the first direction vector of the gathered data, the first direction is vertical with refining direction；
The covariance matrix of the gathered data is calculated, and is chosen in the covariance matrix orthonormalization characteristic vector
The larger preceding t value of characteristic value, composition refines coefficient；
Calculated by the refinement coefficient and obtain refining data.
It is preferred that, the computing unit is suitable to：Will be vectorial based on the first direction vector of the gathered data, it is described
First direction is vertical with refining direction；Gathered data classification is obtained into all kinds of first direction vectors according to identical data direction
Obk_Sx1、Obk_Sx2、…、Obk_Sxi；The vectorial average of Different categories of samples first direction is calculated by all kinds of first direction vectors
Obk_Sm1, Obk_Sm2 ..., the vectorial average Obk_Sm of Obk_Smi and total sample first direction, and and then calculate sample class
Interior scatter matrix Obk_Ss1, Obk_Ss2 ..., scatter matrix Obk_Ssb between Obk_Ssi and sample；Obk_Ssi-1*
The Obk_Ssb corresponding characteristic vector of preceding t eigenvalue of maximum be respectively wi1, wi2 ..., wit, by wi1, wi2 to wit combine
The refinement data are used as into a matrix.
It is preferred that, the data direction of the k-space data include it is following any one or more：Frequency coding direction, phase
Second phase coding direction or above-mentioned multiple directions in position coding direction, channel direction, 3d scannings merge the direction to be formed.
It is preferred that, the data mining method is matrix method of descent.
It is preferred that, column vector direction or row vector direction of the first direction for the gathered data.
By technical scheme provided by the present invention, the convolution kernel chosen in process of reconstruction is gathered for magnetic resonance parallel,
Carry out data and bring up again refining so that the data volume after refinement reduces, and signal characteristic is strengthened, using the convolution kernel after refinement, entered
Row data filling, obtains new convolution kernel and is optimized than original.So noise on the one hand can be removed to a certain degree
Influence so that signal characteristic is strengthened, the accuracy that optimized coefficients are calculated, stability；Further, the choosing of convolution kernel can be caused
Take and be more prone to.
Brief description of the drawings
Fig. 1 is that prior art coil merges coefficient calculating schematic diagram；
Fig. 2 is that magnetic resonance parallel imaging coil of the present invention merges coefficient calculation method flow chart.
Embodiment
In order that the above-mentioned purpose of the present invention, feature, advantage can be more aobvious and understandable, with reference to the accompanying drawings and examples
Embodiment to the present invention is further described.
Merge coefficient calculation method the invention provides a kind of magnetic resonance parallel imaging coil, Fig. 2 is its flow chart, please be joined
See Fig. 2, methods described comprises the following steps：
S101, gathers k-space data, and the k-space data includes gathered data and calibration data；
S102, data mining is carried out to the gathered data and obtains refining data, the gathered data is on direction is refined
Dimension be n, dimension of the refinement data on the refinement direction be t, described t, n are positive integer and t<n；
S103, is calculated by the refinement data and the calibration data and obtains coil merging coefficient.
Step S101 is first carried out, k-space data is acquired, in traditional GRAPPA technologies, it is collected
Data be divided into gathered data and calibration data.Wherein, gathered data is collected by parallel acquisition mode, as shown in figure 1, its
Gathered data is the data of interlacing in k-space, and blank parts are the deficient gathered data not collected, and calibration data is then logical
Cross the data of the k-space central area part collected entirely.When calculating coil merging coefficient, above-mentioned gathered data and calibration
Relation is between data：
Obk_S*Cft=Obk_Sacs [1]
Wherein Obk_S represents gathered data, and Obk_Sacs represents calibration data, and Cft represents to be asked in GRAPPA algorithm
Coil merges coefficient.
As described in background information part, the k-space data collected generally comprises multiple data directions, traditional data side
To for phase-encoding direction and channel direction, in recent years in order to extend the data, frequency coding direction, k-t have also been used
In time orientation, and carry out 3d scannings when second phase encode direction.The data direction of K space data can also be
Above-mentioned one or more directions merge the direction to be formed.For embodiments of the present invention, description afterwards will introduce data side
To number have no effect on the present invention implementation.
Next S102 is performed, data mining is carried out to the gathered data and obtains refining data, the gathered data exists
Refine direction on dimension be n, dimension of the refinement data on the refinement direction be t, described t, n be positive integer and
t<n。
Here, it is necessary to which gathered data Obk_S is refined, refinement data Dbk_S, gathered data Obk_S is obtained and is carried
Relation between refining data Dbk_S is：
Dbk_S=Obk_S*Rft [2]
Wherein, Rft is refinement coefficient.
It is determined that the thought for refining coefficient method is, by vector based on the first direction vector in gathered data Obk_S,
Progress principal component analysis extraction, the correlation gone between vector, first direction vector here can be in gathered data matrix
The column vector of each row or the row vector per a line.Here the method selected be matrix dimensionality reduction mathematical method, such as PCA,
A kind of arbitrary method in the methods such as KLT, LDA.
KLT algorithms are characterized one of conventional algorithm of extraction comparison [referring to document：1.M.Turk and A.Pentland,
“Eigenfaces for recognition,”J.Cogn.Neurosci.3,71–86(1991).2.R.Everson and L.
" Sirovich, Karhunen-Loeve procedure for gappy data " Vol.12, No.8/August 1995/
J.Opt.Soc.Am.A]。
The dimensionality reduction of data matrix is acquired by taking KLT algorithms as an example, and then when determining to refine coefficients R ft, its implementation process
For using gathered data Obk_S column vector, as extracting object, the column vector equivalent in Fig. 1, count by corresponding row black
According to.Gathered data Obk_S covariance matrix C_Obk_S is tried to achieve first, then tries to achieve C_Obk_S orthonormalization Characteristic Vectors
Q (it is assumed that common n) is measured, selected characteristic is worth larger preceding t (t<N) individual q, constitutes Rft, and this is to refine coefficient.In said process, it is
It is vectorial based on column vector, here equally can will be vectorial based on row vector.Vectorial based on column vector
When, its direction vertical with column vector is designated as extracting direction (line direction), and gathered data has n dimension on direction is extracted, by
There is t dimension in refining coefficients R ft, the extraction data obtained after extraction are also t dimension on direction (line direction) is extracted
Degree, wherein the t chosen<n.
If by taking LDA algorithm as an example, the specific implementation process of the technology of the present invention is：First by the row in gathered data Obk_S
Vector is classified, according to data direction (such as：Frequency coding direction and channel direction) identical vector be designated as a class
Obk_Sx1, Obk_Sx2 ..., Obk_Sxi, the common c class of acquisition matrix, N number of vector.All column vectors of i-th class are represented
For Ri, its common Ni column vector.Wherein, c, N, i, Ni are positive integer.
Remember Different categories of samples mean vector be Obk_Sm1, Obk_Sm2 ..., Obk_Smi, its calculating process is：
Total sample average Obk_Sm, its calculating process is：
Matrix within samples Obk_Ss1, Obk_Ss2 ..., Obk_Ssi, its calculating process is：
Matrix between samples Obk_Ssb, its calculating process is：
It is above-mentioned it is various in, x represents a column vector in the i-th class column vector Ri, c, N, i, Ni with providing identical before,
(…)TIt is expressed as the inversion of matrix.
It is matrix Obk_Ssi to remember wi1, wi2 ... wit-1* the Obk_Ssb corresponding characteristic vector of preceding t eigenvalue of maximum,
Set of eigenvectors one matrix of charge-coupled synthesis that all classes are tried to achieve is to refine data Dbk_S, and refinement can be tried to achieve using formula [2]
Coefficients R ft.
It is vectorial based on column vector is chosen in above-mentioned steps, with KLT algorithm identicals be here can also will row to
It is vectorial based on amount.In the vector based on column vector, line direction is to refine direction, the data on direction is refined
Dimension is n, and the data dimension of the refinement data after refinement is t, t<n.
Step S103 is finally performed, is calculated by the refinement data and the calibration data and obtains coil merging coefficient.
By the refinement of previous step, the final coil merging coefficient base group that calculates includes refinement data Dbk_S and calibration number
According to this two groups of data of Obk_Sacs；
It refines data Dbk_S and calibration data Obk_Sacs corresponding relations are：
Dbk_S*Cft_new=Obk_Sacs [7]
Here change to solve Cft_new process.
According to the step S101 gathered data Obk_S collected and calibration data Obk_Sacs, and respective formula [2]
[7], it is known that：
Obk_S*Rft*Cft_new=Obk_Sacs [8]
In order to calculate simplicity, it is reduced to here：
Cft=Rft*Cft_new [9]
So, then：
Obk_S*Cft=Obk_Sacs [10]
Cft_new is the coil merging coefficient that technical solution of the present invention is obtained in above-mentioned formula, fills up and obtains complete k skies
Between process be exactly to utilize coil to merge coefficient Cft_new and gathered data Obk_S to carry out.
For the description of technical solution of the present invention more than, due in step s 102, passing through the refinement of data
Process, the dimension for refining the merging coefficient of the coil corresponding to data Dbk_S Cft_new is accordingly under control, so in step
When S101 selects gathered data, it can simply select to be handled than larger gathered data, it is not necessary to according to data
Situation converts the size of gathered data.I.e. so that selection gathered data is simpler, it is illustrated below：Common method is chosen
Gathered data, it is assumed that include two data directions for gathered data, be nx*ny (such as 3*4) size；Because calibration data amount has
Nx*ny is unsuitable excessive in limit, fit procedure；But then, it is intended that nx*ny is sufficiently large, as far as possible many information are covered,
So need the size of balance convolution kernel.So using this method, a nx*ny (such as 30*4) can be chosen than larger one
Gathered data, then using the method refined, information useful in nx*ny is extracted and (reduces nx*ny big to 3*4 data
It is small), make it in extractive process, the refinement data finally given diminish.
The present invention merges in above-mentioned magnetic resonance parallel imaging coil on the basis of coefficient calculation method, additionally provides a kind of magnetic
Resonate parallel imaging method, including merges the coil conjunction that coefficient calculation method calculating is obtained by above-mentioned magnetic resonance parallel imaging coil
And coefficient, and gathered data carry out parallel acceleration data reconstruction, fill up and obtain complete k-space data, afterwards by k-space number
According to image area is transformed to, MRI is obtained.
The present invention additionally provides a kind of magnetic resonance parallel imaging device corresponding to MR imaging method, including：
Collecting unit, suitable for collection k-space data, the k-space data includes gathered data and calibration data；
Computing unit, obtains refining data, the gathered data is being carried suitable for carrying out the gathered data data mining
It is n to refine the dimension on direction, and dimension of the refinement data on the refinement direction is t, and described t, n are positive integer and t<
N, and coil merging coefficient is obtained by the refinement data and calibration data calculating；
Shim, suitable for merging coefficient and the gathered data according to the coil, fills up and obtains complete k-space number
According to；
Imaging unit, image is obtained suitable for the complete k-space data is transformed into image area.
Wherein, computing unit is suitable to：By vector, the first party based on the first direction vector of the gathered data
To with refine direction it is vertical；The covariance matrix of the gathered data is calculated, and chooses the covariance matrix orthonormalization
The larger preceding t value of characteristic value in characteristic vector, composition refines coefficient；Calculated by the refinement coefficient and obtain refining data.
Optionally, computing unit is suitable to：By vector, described first based on the first direction vector of the gathered data
Direction is vertical with refining direction；Gathered data classification is obtained into all kinds of first direction vector Obk_ according to identical data direction
Sx1、Obk_Sx2、…、Obk_Sxi；The vectorial average Obk_ of Different categories of samples first direction is calculated by all kinds of first direction vectors
Sm1, Obk_Sm2 ..., the vectorial average Obk_Sm of Obk_Smi and total sample first direction, and and then calculate in sample class from
Scatter Matrix Obk_Ss1, Obk_Ss2 ..., scatter matrix Obk_Ssb between Obk_Ssi and sample；Obk_Ssi-1*Obk_
The Ssb corresponding characteristic vector of preceding t eigenvalue of maximum be respectively wi1, wi2 ..., wit, wi1, wi2 to wit are combined into one
Individual matrix is used as the refinement data.
The specific implementation process of above-mentioned MR imaging method and MR imaging apparatus refers to magnetic resonance parallel imaging
Coil merges the implementation process of coefficient calculation method, no longer repeats one by one here.
Although the present invention is disclosed as above with preferred embodiment, so it is not limited to the present invention, any this area skill
Art personnel, without departing from the spirit and scope of the present invention, when a little modification can be made and perfect, therefore the protection model of the present invention
Enclose when by being defined that claims are defined.
Claims (5)
1. a kind of magnetic resonance parallel imaging device, it is characterised in that including：
Collecting unit, suitable for collection k-space data, the k-space data includes gathered data and calibration data；
Computing unit, obtains refining data, the gathered data is in refinement side suitable for carrying out the gathered data data mining
Upward dimension is n, and dimension of the refinement data on the refinement direction is t, and described t, n are positive integer and t<N, and
Calculated by the refinement data and the calibration data and obtain coil merging coefficient；
Shim, suitable for merging coefficient and the gathered data according to the coil, fills up and obtains complete k-space data；
Imaging unit, image is obtained suitable for the complete k-space data is transformed into image area；
The computing unit is suitable to：
By vector based on the first direction vector of the gathered data, the first direction is vertical with refining direction；
The covariance matrix of the gathered data is calculated, and chooses feature in the covariance matrix orthonormalization characteristic vector
The larger preceding t value of value, composition refines coefficient；
Calculated by the refinement coefficient and obtain refining data..
2. magnetic resonance parallel imaging device as claimed in claim 1, it is characterised in that the computing unit is suitable to：
By vector based on the first direction vector of the gathered data, the first direction is vertical with refining direction；
According to identical data direction by the gathered data classification obtain all kinds of first direction vector Obk_Sx1, Obk_Sx2 ...,
Obk_Sxi；
By all kinds of first directions vector calculate the vectorial average Obk_Sm1 of Different categories of samples first direction, Obk_Sm2 ..., Obk_
The vectorial average Obk_Sm of Smi and total sample first direction, and and then calculate matrix within samples Obk_Ss1,
Obk_Ss2 ..., scatter matrix Obk_Ssb between Obk_Ssi and sample；
Obk_Ssi-1* the Obk_Ssb corresponding characteristic vector of preceding t eigenvalue of maximum be respectively wi1, wi2 ..., wit, will
Wi1, wi2 are combined into a matrix as the refinement data to wit.
3. magnetic resonance parallel imaging device as claimed in claim 1, it is characterised in that the data direction of the k-space data
Including it is following any one or more：Second phase in frequency coding direction, phase-encoding direction, channel direction, 3d scannings
Coding direction or above-mentioned multiple directions merge the direction to be formed.
4. magnetic resonance parallel imaging device as claimed in claim 1, it is characterised in that the data mining method drops for matrix
Dimension method.
5. magnetic resonance parallel imaging device as claimed in claim 1, it is characterised in that the first direction is the collection number
According to column vector direction or row vector direction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710368811.0A CN107092805B (en) | 2014-01-09 | 2014-01-09 | Magnetic resonance parallel imaging device |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710368811.0A CN107092805B (en) | 2014-01-09 | 2014-01-09 | Magnetic resonance parallel imaging device |
| CN201410010303.1A CN104778341B (en) | 2014-01-09 | 2014-01-09 | Magnetic resonance coil merges coefficient calculation method, MR imaging method and its device |
Related Parent Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410010303.1A Division CN104778341B (en) | 2014-01-09 | 2014-01-09 | Magnetic resonance coil merges coefficient calculation method, MR imaging method and its device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107092805A true CN107092805A (en) | 2017-08-25 |
| CN107092805B CN107092805B (en) | 2020-08-04 |
Family
ID=53619802
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410010303.1A Active CN104778341B (en) | 2014-01-09 | 2014-01-09 | Magnetic resonance coil merges coefficient calculation method, MR imaging method and its device |
| CN201710368811.0A Active CN107092805B (en) | 2014-01-09 | 2014-01-09 | Magnetic resonance parallel imaging device |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410010303.1A Active CN104778341B (en) | 2014-01-09 | 2014-01-09 | Magnetic resonance coil merges coefficient calculation method, MR imaging method and its device |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN104778341B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107582057A (en) * | 2017-09-30 | 2018-01-16 | 上海联影医疗科技有限公司 | MR imaging method and device |
| CN110146835A (en) * | 2019-05-22 | 2019-08-20 | 山东颐邦齐鲁医生集团管理有限公司 | A kind of auto-navigation MR image reconstruction method and device based on parallel imaging |
| CN112557980A (en) * | 2020-11-02 | 2021-03-26 | 上海东软医疗科技有限公司 | Magnetic resonance image correction method, magnetic resonance image correction device, medium, and electronic apparatus |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111582453B (en) * | 2020-05-09 | 2023-10-27 | 北京百度网讯科技有限公司 | Method and device for generating neural network model |
| CN111951344B (en) * | 2020-08-09 | 2022-08-02 | 昆明理工大学 | Magnetic resonance image reconstruction method based on cascade parallel convolution network |
| WO2022218438A1 (en) * | 2021-04-16 | 2022-10-20 | Shanghai United Imaging Healthcare Co., Ltd. | Calibration methods and systems for imaging field |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060273792A1 (en) * | 2005-04-25 | 2006-12-07 | Kholmovski Evgueni G | Systems and methods for image reconstruction of sensitivity encoded MRI data |
| WO2008130679A1 (en) * | 2007-04-18 | 2008-10-30 | Koninklijke Philips Electronics N.V. | Reconstruction of a magnetic resonance image in image space using basis functions (rib) for partially parallel imaging |
| US20100237864A1 (en) * | 2009-03-23 | 2010-09-23 | Alto Stemmer | Magnetic resonance method and apparatus to generate an image using a parallel acquisition technique |
| CN103185878A (en) * | 2011-12-27 | 2013-07-03 | 上海联影医疗科技有限公司 | Magnetic resonance parallel image acquisition and image reconstruction method |
| CN103430038A (en) * | 2011-03-17 | 2013-12-04 | 皇家飞利浦有限公司 | A MRI method of faster channel-by-channel reconstruction without image degradation |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5887074A (en) * | 1996-12-13 | 1999-03-23 | Siemens Corporate Research, Inc. | Local principal component based method for detecting activation signals in functional MR images |
| WO2005047919A1 (en) * | 2003-11-12 | 2005-05-26 | Invivo Corporation | Method for generating fast magnetic resonance images |
| EP2063392A1 (en) * | 2007-11-24 | 2009-05-27 | Barco NV | Image processing of medical images |
| CN102366323B (en) * | 2011-09-30 | 2013-09-11 | 中国科学院自动化研究所 | Detection method for causal connection strength of magnetic resonance brain imaging based on PCA (Principal component analysis) and GCA (Granger causality analysis) |
| CN102749600B (en) * | 2012-05-30 | 2014-12-24 | 苏州朗润医疗***有限公司 | Synthetic method of magnetic resonance multi-channel image |
-
2014
- 2014-01-09 CN CN201410010303.1A patent/CN104778341B/en active Active
- 2014-01-09 CN CN201710368811.0A patent/CN107092805B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060273792A1 (en) * | 2005-04-25 | 2006-12-07 | Kholmovski Evgueni G | Systems and methods for image reconstruction of sensitivity encoded MRI data |
| WO2008130679A1 (en) * | 2007-04-18 | 2008-10-30 | Koninklijke Philips Electronics N.V. | Reconstruction of a magnetic resonance image in image space using basis functions (rib) for partially parallel imaging |
| US20100237864A1 (en) * | 2009-03-23 | 2010-09-23 | Alto Stemmer | Magnetic resonance method and apparatus to generate an image using a parallel acquisition technique |
| CN103430038A (en) * | 2011-03-17 | 2013-12-04 | 皇家飞利浦有限公司 | A MRI method of faster channel-by-channel reconstruction without image degradation |
| CN103185878A (en) * | 2011-12-27 | 2013-07-03 | 上海联影医疗科技有限公司 | Magnetic resonance parallel image acquisition and image reconstruction method |
Non-Patent Citations (2)
| Title |
|---|
| ASHISH RAJ ETAL.: "A Maximum Likelihood Approach to Parallel Imaging With Coil Sensitivity Noise", 《IEEE TRANSACTIONS ON MEDICAL IMAGING》 * |
| 李逸明等: "磁共振并行成像线圈敏感度计算方法的比较分析", 《中国医疗器械杂志研究与论著》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107582057A (en) * | 2017-09-30 | 2018-01-16 | 上海联影医疗科技有限公司 | MR imaging method and device |
| CN107582057B (en) * | 2017-09-30 | 2020-10-27 | 上海联影医疗科技有限公司 | Magnetic resonance imaging method and device |
| CN110146835A (en) * | 2019-05-22 | 2019-08-20 | 山东颐邦齐鲁医生集团管理有限公司 | A kind of auto-navigation MR image reconstruction method and device based on parallel imaging |
| CN110146835B (en) * | 2019-05-22 | 2021-09-07 | 山东颐邦齐鲁医生集团管理有限公司 | Self-navigation magnetic resonance image reconstruction method and device based on parallel imaging |
| CN112557980A (en) * | 2020-11-02 | 2021-03-26 | 上海东软医疗科技有限公司 | Magnetic resonance image correction method, magnetic resonance image correction device, medium, and electronic apparatus |
| CN112557980B (en) * | 2020-11-02 | 2022-05-03 | 上海东软医疗科技有限公司 | Magnetic resonance image correction method, magnetic resonance image correction device, medium, and electronic apparatus |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104778341A (en) | 2015-07-15 |
| CN107092805B (en) | 2020-08-04 |
| CN104778341B (en) | 2017-08-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104778341B (en) | Magnetic resonance coil merges coefficient calculation method, MR imaging method and its device | |
| CN110414377B (en) | Remote sensing image scene classification method based on scale attention network | |
| CN110969250B (en) | Neural network training method and device | |
| CN108896943B (en) | Magnetic resonance quantitative imaging method and device | |
| Myllymäki et al. | GET: Global envelopes in R | |
| CN106796716B (en) | For providing the device and method of super-resolution for low-resolution image | |
| CN104020430B (en) | Correction method and system for movement artifacts of magnetic resonance imaging | |
| KR101599219B1 (en) | system and method for automatic registration of anatomic points in 3d medical images | |
| Zhang et al. | LU-NET: An improved U-Net for ventricular segmentation | |
| CN108647741A (en) | A kind of image classification method and system based on transfer learning | |
| CN102651124B (en) | Image fusion method based on redundant dictionary sparse representation and evaluation index | |
| CN109360147B (en) | Multispectral image super-resolution reconstruction method based on color image fusion | |
| CN113283419B (en) | Convolutional neural network pointer instrument image reading identification method based on attention | |
| CN108805061A (en) | Hyperspectral image classification method based on local auto-adaptive discriminant analysis | |
| CN102422324A (en) | Age estimation device, method, and program | |
| CN107729926A (en) | A kind of data amplification method based on higher dimensional space conversion, mechanical recognition system | |
| KR102561214B1 (en) | A method and apparatus for image segmentation using global attention | |
| CN106650572A (en) | Method for assessing quality of fingerprint image | |
| CN110853048A (en) | MRI image segmentation method, device and storage medium based on rough training and fine training | |
| CN109613611A (en) | The determination method and system of input-to-state stabilization for earthquake-resistant structure time-history analysis | |
| CN109859189A (en) | A kind of age estimation method based on deep learning | |
| CN101980251A (en) | Remote sensing classification method for binary tree multi-category support vector machines | |
| KR102604216B1 (en) | A method and apparatus for image segmentation using global attention-based convolutional network | |
| CN109829494A (en) | A kind of clustering ensemble method based on weighting similarity measurement | |
| CN108460400A (en) | A kind of hyperspectral image classification method of combination various features information |
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 | ||
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 201807 Shanghai city Jiading District Industrial Zone Jiading Road No. 2258 Patentee after: Shanghai Lianying Medical Technology Co., Ltd Address before: 201807 Shanghai city Jiading District Industrial Zone Jiading Road No. 2258 Patentee before: SHANGHAI UNITED IMAGING HEALTHCARE Co.,Ltd. |