CN114325524B - Magnetic resonance image reconstruction method, device, system and storage medium - Google Patents
Magnetic resonance image reconstruction method, device, system and storage medium Download PDFInfo
- Publication number
- CN114325524B CN114325524B CN202011051798.4A CN202011051798A CN114325524B CN 114325524 B CN114325524 B CN 114325524B CN 202011051798 A CN202011051798 A CN 202011051798A CN 114325524 B CN114325524 B CN 114325524B
- Authority
- CN
- China
- Prior art keywords
- fitting
- sampling points
- sampling
- mode
- space
- 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.)
- Active
Links
Classifications
-
- 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
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/30—Assessment of water resources
Abstract
The embodiment of the invention discloses a magnetic resonance image reconstruction method, a device, a system and a storage medium, wherein the method comprises the following steps: acquiring a K space to be processed, wherein the K space comprises a plurality of sampling points and a plurality of non-sampling points; determining a fitting mode corresponding to a plurality of non-sampling points, wherein the fitting mode is a pattern formed by sampling points within a set range from each non-sampling point; setting a fitting mode of each non-sampling point according to fitting modes corresponding to the non-sampling points; obtaining a fitting result of each non-sampling point by using a fitting mode of each non-sampling point, wherein K space data of the plurality of sampling points and the fitting result of the non-sampling points form a fitting K space; reconstructing the fit K-space to acquire a magnetic resonance image. The method solves the problem that the reconstruction speed of the magnetic resonance image cannot be obviously improved due to the fact that the fitting speed of the non-sampling points is low.
Description
Technical Field
The embodiment of the invention relates to the field of medical images, in particular to a method, a device, a system and a storage medium for reconstructing a magnetic resonance image.
Background
The magnetic resonance imaging system comprises a main magnet, a gradient coil, a radio frequency transmitting coil, a radio frequency receiving coil, and an image reconstruction unit. The spin of hydrogen nuclei in the human body can be equivalently a small magnetic needle. In the strong magnetic field provided by the main magnet, hydrogen nuclei are converted from a disordered thermal equilibrium state into a partially forward and a partially reverse state, and the difference between the hydrogen nuclei and the main magnetic field direction forms a net magnetization vector. The hydrogen nuclei precess around the main magnetic field, with the precession frequency being proportional to the magnetic field strength. The gradient unit generates a magnetic field with a strength that varies with spatial position for spatial encoding of the signal. The radio frequency transmitting coil is used for turning hydrogen nuclei from the direction of the main magnetic field to a transverse plane, precessing around the main magnetic field, and finally inducing current signals in the radio frequency receiving coil so as to obtain magnetic resonance data. The image reconstruction unit is used for reconstructing images of the magnetic resonance data to obtain magnetic resonance images.
Because fully acquired magnetic resonance data is particularly huge, in order to increase the acquisition speed of the magnetic resonance data, an undersampling mode is generally adopted to acquire the magnetic resonance data. After undersampled magnetic resonance data is obtained, the image reconstruction unit needs to fit all the undersampled points in the undersampled magnetic resonance data, and then performs image reconstruction on the fitted magnetic resonance data to obtain a magnetic resonance image. Since the more the number of non-sampled points, the more time it takes for non-sampled point fitting, and the more time it takes for undersampling fitting, the slower the image reconstruction speed of the magnetic resonance data, and the more difficult it is to substantially increase.
In summary, the prior art has the problem that the magnetic resonance image reconstruction speed cannot be remarkably improved due to the fact that the fitting speed of the non-sampling points is low.
Disclosure of Invention
The embodiment of the invention provides a magnetic resonance image reconstruction method, a device, a system and a storage medium, which solve the problem that the reconstruction speed of a magnetic resonance image cannot be obviously improved due to the fact that the fitting speed of non-sampling points is low.
In a first aspect, an embodiment of the present invention provides a magnetic resonance image reconstruction method, including:
acquiring a K space to be processed, wherein the K space comprises a plurality of sampling points and non-sampling points;
determining a fitting mode corresponding to a plurality of non-sampling points, wherein the fitting mode is a pattern formed by sampling points within a set range from each non-sampling point;
setting a fitting mode of each non-sampling point according to fitting modes corresponding to the non-sampling points;
obtaining a fitting result of each non-sampling point by using a fitting mode of each non-sampling point, wherein K space data of the plurality of sampling points and the fitting result of the non-sampling points form a fitting K space;
reconstructing the fit K-space to acquire a magnetic resonance image.
In a second aspect, an embodiment of the present invention further provides a magnetic resonance image reconstruction apparatus, including:
The acquisition module is used for acquiring a K space to be processed, wherein the K space comprises a plurality of sampling points and non-sampling points;
the fitting mode determining module is used for determining fitting modes corresponding to a plurality of non-sampling points, wherein the fitting modes are patterns formed by sampling points within a set range from each non-sampling point;
the fitting mode determining module is used for setting the fitting mode of each non-sampling point according to the fitting modes corresponding to the non-sampling points;
the fitting module is used for obtaining a fitting result of each non-sampling point by using a fitting mode of each non-sampling point, and K space data of the plurality of sampling points and the fitting result of the non-sampling points form a fitting K space;
and the reconstruction module is used for reconstructing the fitting K space so as to acquire a magnetic resonance image.
In a third aspect, an embodiment of the present invention further provides a magnetic resonance system, including:
the radio frequency transmitting coil is used for transmitting radio frequency pulses to a scanning part of a target object so as to excite nuclear spins of the scanning part;
gradient coils for applying slice selection gradient fields, phase encoding gradient fields, and frequency encoding gradient fields to the scan site to generate echo signals;
a radio frequency receiving coil for receiving the echo signals to form magnetic resonance scan data;
The processor is used for acquiring a K space to be processed, wherein the K space comprises a plurality of sampling points and non-sampling points; determining a fitting mode corresponding to a plurality of non-sampling points, wherein the fitting mode is a pattern formed by sampling points within a set range from each non-sampling point; setting a fitting mode of each non-sampling point according to fitting modes corresponding to the non-sampling points; obtaining a fitting result of each non-sampling point by using a fitting mode of each non-sampling point, wherein K space data of the plurality of sampling points and the fitting result of the non-sampling points form a fitting K space; reconstructing the fit K-space to acquire a magnetic resonance image.
In a fourth aspect, embodiments of the present invention also provide a storage medium containing computer-executable instructions, which when executed by a computer processor, are configured to perform the magnetic resonance image reconstruction method of any of the embodiments.
Compared with the prior art, the technical scheme of the magnetic resonance image reconstruction method provided by the embodiment of the invention determines the fitting modes corresponding to the plurality of non-sampling points, sets the fitting mode of each non-sampling point according to the fitting modes corresponding to the plurality of non-sampling points, and acquires the fitting result of each non-sampling point by utilizing the fitting mode of each non-sampling point. By matching different fitting modes for different fitting modes, the fitting time of the non-sampling points of the different fitting modes is reduced, so that the fitting time of all the non-sampling points in the whole K space is reduced, the fitting speed of the non-sampling points in the K space is improved, and the technical effect of improving the reconstruction speed of the magnetic resonance image is achieved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a magnetic resonance image reconstruction method according to a first embodiment of the present invention;
FIG. 2 is a schematic diagram of fitting non-sampled points according to a first embodiment of the present invention;
FIG. 3A is a schematic view of a fitting pattern according to a first embodiment of the present invention;
FIG. 3B is a schematic view of another fitting pattern provided in accordance with an embodiment of the present invention;
FIG. 3C is a schematic view of another fitting pattern provided in accordance with an embodiment of the present invention;
fig. 4 is a block diagram of a magnetic resonance image reconstruction device according to a second embodiment of the present invention;
fig. 5 is a block diagram of a magnetic resonance system according to a third embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described by means of implementation examples with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1
Fig. 1 is a flowchart of a magnetic resonance image reconstruction method according to an embodiment of the present invention. The invention provides a magnetic resonance image reconstruction method aiming at the problem of low reconstruction speed under the condition of multichannel coil receiving when magnetic resonance imaging is carried out rapidly, in particular to 3D high-definition imaging. The method can be implemented by the magnetic resonance image reconstruction device provided by the embodiment of the invention, and the device can be implemented in a software and/or hardware mode and is configured to be applied in a processor of the processor. The method specifically comprises the following steps:
s101, acquiring a K space to be processed, wherein the K space comprises a plurality of sampling points and a plurality of non-sampling points.
The K space to be processed is the K space currently used for image reconstruction, and the K space comprises a plurality of sampling points and at least two non-sampling points.
S102, determining a fitting mode corresponding to a plurality of non-sampling points, wherein the fitting mode is a pattern formed by sampling points within a set range from each non-sampling point. That is, the corresponding fitting pattern is determined by the pattern formed by the sampling points within the set range from each sampling point. As shown in fig. 2, the non-sampled points can be obtained from the sampled points in the vicinity thereof by linear fitting, and the fitting formula is as follows:
S mn ＝∑ i,j C i,j S i,j (1)
Wherein S is mn Representing the value of the non-sampling point obtained by fitting, wherein the non-sampling point corresponds to the m-th channel and the K space coordinate is n; c (C) i,j As the weight coefficient, S i,j Fitting values of sampling points within a range for non-sampling points, i representing the channel of sampling points for fitting, j representing the channel for fittingIs not equal to n, and the sampling points used for fitting are near the K-space coordinate n. Weight coefficient C i,j The calibration data may be obtained by calculation, and the present embodiment is not described herein. In this embodiment, m and i are both positive integers, m=i for the case of single channel acquisition; for the case of multi-channel acquisition, i can take any channel value of 1, 2, 3, etc., and m is one of the multiple channels. In one embodiment, the coordinates of the K space may be any value from-127 to 128 in the phase encoding direction, -127. Ltoreq.j. Ltoreq.128, and-127. Ltoreq.n. Ltoreq.128.
It can be seen that when the number and distribution of sampling points around the non-sampling points are different, the weight coefficient matrix is also different, and the concept of fitting mode is introduced for this embodiment. The number and distribution of each sampling point is made to correspond to a fitting mode, namely, each weight coefficient matrix is made to correspond to a fitting mode. The non-sampled points (shaded points) in fig. 3A, 3B and 3C correspond to three different fitting patterns, respectively.
It will be appreciated that if the current sampling mode of the K-space is interlaced, the non-sampled points located in the middle portion of the K-space all correspond to the same fitting pattern, i.e., the fitting pattern in fig. 3A. If the current K-space data is sampled in a random manner, see fig. 3B and 3C, then the fit patterns corresponding to different non-sampled points of the K-space are typically different, or each fit pattern corresponds to only one or a small number of non-sampled points.
In one embodiment, if the distribution of sampling points within the fitting range of the non-sampling points is regular, the non-sampling points correspond to the first fitting pattern, see fig. 3A; if the distribution of the sampling points within the fitting range of the non-sampling points is irregular, the non-sampling points correspond to the second fitting pattern, see fig. 3B and 3C. It is understood that the position points of the K-space contained in the first fitting pattern are different from the position points of the K-space contained in the second fitting pattern. The first fitting pattern includes a different number of sampling points than the second fitting pattern.
S103, setting a fitting mode of each non-sampling point according to fitting modes corresponding to the non-sampling points. In this embodiment, the fitting manner of the non-sampling points may also be expressed as a reconstruction method of the non-sampling points, one reconstruction method is selected for each non-sampling point from at least two different reconstruction methods according to fitting patterns corresponding to the non-sampling points, and the non-sampling point recovery process of the whole K space uses a hybrid reconstruction method, that is, two or more reconstruction methods.
According to the magnitude relation between the first fitting time and the second fitting time corresponding to each fitting mode, the fitting mode corresponding to each fitting mode is determined, so that the fitting time of all the non-sampling points corresponding to each fitting mode is the shortest.
The K-space fit corresponds to equation 1, and without loss of generality, can be represented as a convolution operation as follows:
wherein S is m Fitting result S for a plurality of non-sampled points mn Matrix formed, C' i Is C i,j Matrix obtained by coordinate inversion, S i Is S i,j The K-space data/matrix is composed,is a convolution operation.
According to the mathematical principle, the image domain fitting (image domain product fitting) corresponds to formula 2, and the convolution operation can be rewritten as:
S m ＝FFT(∑ i W i ·I i ) (3)
wherein W is i ＝IFFT(C' i )，I i ＝IFFT(S i ) FFT means Fourier transform, IFFT means Fourier inverse transform, and ". Cndot." means point multiplication operation.
It is verified that, for K-space data with uniform undersampling, such as that shown in fig. 3A, sampling points and non-sampling points are distributed in an interlaced manner along a phase encoding direction (an abscissa direction in the drawing), a complete line is formed corresponding to a plurality of sampling points with the same phase encoding, a complete line is formed corresponding to a plurality of non-sampling points with the same phase encoding, and the operation speed of formula 3 is faster than that of formula 2, but in the case that a data matrix contained in K-space is smaller or the number of non-sampling points to be fitted is small, or in the case of non-uniform sampling, as shown in fig. 3B and 3C, the operation speed of formula 2 is faster. It can be seen that for any non-sampling point, the two fitting times corresponding to the non-sampling points are different, so that the fitting modes corresponding to the non-sampling points are classified, the non-sampling points belonging to the same fitting mode are classified into one type, and the non-sampling points with the same fitting mode are counted; and calculating a first fitting time corresponding to the non-sampling points in the same fitting mode in a K space fitting mode and a second fitting time corresponding to the non-sampling points in the same fitting mode in an image domain mode, and then selecting the fitting mode corresponding to the small one of the first fitting time and the second fitting time as the fitting mode of each non-sampling point.
In order to further improve the fitting speed of the non-sampling points, if the sampling point distribution rule corresponding to the non-sampling points is detected, namely that the non-sampling points correspond to the first fitting mode is detected, the fitting mode is set to be an image domain fitting mode. If irregular distribution of sampling points corresponding to the non-sampling points is detected, namely, a second fitting mode corresponding to the non-sampling points is detected, and a fitting mode corresponding to each non-sampling point is determined according to the size relation between the number of the sampling points corresponding to the fitting mode and a preset threshold value. Specifically, if the number of non-sampled points in the same fitting mode is smaller than a set threshold, it is known that the fitting time of the corresponding K-space fitting mode is smaller than the fitting time of the corresponding image domain, and then the fitting mode of the non-sampled points in the fitting mode is set as a K-space fitting mode; if the number of the non-sampling points in the same fitting mode is larger than a set threshold, the fitting time of the corresponding K space fitting mode is larger than or equal to the fitting time of the corresponding image domain, and at the moment, the fitting mode of the non-sampling points in the fitting mode is set as the image domain fitting mode.
In one embodiment, the preset threshold is determined by: according to a reconstruction computer with a specific configuration and a specific K space size, different thresholds are selected from small to large, the thresholds are equal to the number H of the non-sampling points of the K space to be fitted, and the time required for fitting the H non-sampling points by using the K space fitting and the image domain respectively is calculated. When the time required for fitting using K-space is equal to the time required for fitting of the image domain, the threshold in this case is the set threshold. The set threshold may be stored within the reconstruction code.
The phase-encoding lines or position points that fill the K-space peripheral region mainly determine the anatomical details of the image, considering that the phase-encoding lines or position points in the K-space central region mainly determine the contrast of the image. The number of sampling points of the K space to be processed in the central area of the K space is larger than that of sampling points of the peripheral area of the K space. In an embodiment, a K-space fit or K-space based reconstruction is selected for the fit/reconstruction method of the non-sampled points near the central region of K-space, i.e. the non-sampled points at low frequencies select the K-space fit; and selecting an image domain fitting mode or an image space-based reconstruction mode for the fitting mode of the non-sampling points of the K space peripheral region, namely selecting an image domain fitting mode for the non-sampling points with high frequency.
It should be noted that, in this embodiment, according to whether the distribution of sampling points in the fitting range of the non-sampling points is regular, the fitting modes of the non-sampling points may be divided into a first fitting mode and a second fitting mode, specifically, if the distribution of sampling points is regular, the fitting modes of the non-sampling points corresponding to the non-sampling points are the first fitting mode, and if the distribution of sampling points is irregular, the fitting modes of the non-sampling points corresponding to the non-sampling points are the second fitting mode, then the second fitting mode is subdivided according to the number and distribution of the sampling points, and the subdivided fitting modes may be a 2A fitting mode, a 2B fitting mode, and so on.
Wherein the magnitude of the set threshold is related to the image processing parameters of the processor. If the configuration of the image processing parameters of the processor is higher, the image processing speed is higher, and the fitting mode with a little more fitting points is fitted by adopting a K space fitting mode, the set threshold will be higher if less effort is required, otherwise, if the configuration of the performance parameters of the processor is lower, the image processing speed is slower, and the fitting speed is rapidly reduced when the fitting mode with a little more fitting points is fitted by adopting a K space fitting mode, so the set threshold will be lower.
S104, acquiring a fitting result of each non-sampling point by using a fitting mode of each non-sampling point, wherein K space data of a plurality of sampling points and the fitting result of the non-sampling points form a fitting K space.
Fitting each non-sampling point by using a fitting mode of each non-sampling point so as to obtain a fitting result of each non-sampling point. Fitting results of all the non-sampling points and K space data of a plurality of sampling points form a fitting space.
In some embodiments, after determining the fitting manner corresponding to the non-sampling points of each fitting mode, fitting all the non-sampling points corresponding to each fitting mode by adopting the fitting manner corresponding to each fitting mode, so as to obtain the fitting result of all the non-sampling points corresponding to the same fitting mode at one time. After the fitting operation is performed on all the non-sampled points corresponding to all the fitting modes, the fitting result of all the non-sampled points in the K space can be obtained.
In one embodiment, all fitting modes are traversed, the fitting point number of the current fitting mode is determined, the fitting mode corresponding to the current fitting mode is determined according to the size relation between the fitting point number and a set threshold value, and fitting is carried out on all non-sampled points corresponding to the current fitting mode by adopting the fitting mode, so that the fitting result of all non-sampled points corresponding to the current fitting mode is obtained.
For the image domain fitting mode, the direct fitting result is image domain data, the fitting result is required to be transformed into K space to obtain a transformation result, fitting data of one or more non-sampling points corresponding to the corresponding fitting mode are extracted from the transformation result, and the extracted fitting data are respectively filled into the corresponding non-sampling points of the K space to update the fitting result corresponding to the current fitting mode.
S105, reconstructing the fitting K space to acquire a magnetic resonance image.
After the fitted K space is obtained, the fitted K space is subjected to image reconstruction to obtain a magnetic resonance image.
Compared with the prior art, the technical scheme of the magnetic resonance image reconstruction method provided by the embodiment of the invention determines the fitting modes corresponding to the plurality of non-sampling points, sets the fitting mode of each non-sampling point according to the fitting modes corresponding to the plurality of non-sampling points, and acquires the fitting result of each non-sampling point by utilizing the fitting mode of each non-sampling point. By matching different fitting modes for different fitting modes, the fitting time of the non-sampling points of the different fitting modes is reduced, so that the fitting time of all the non-sampling points in the whole K space is reduced, the fitting speed of the non-sampling points in the K space is improved, and the technical effect of improving the reconstruction speed of the magnetic resonance image is achieved.
Example two
Fig. 4 is a block diagram of a magnetic resonance image reconstruction apparatus according to an embodiment of the present invention. The apparatus is used for executing the magnetic resonance image reconstruction method provided by any of the above embodiments, and the apparatus may be implemented in software or hardware. The device comprises:
an obtaining module 11, configured to obtain a K space to be processed, where the K space includes a plurality of sampling points and non-sampling points;
a fitting pattern determining module 12, configured to determine fitting patterns corresponding to a plurality of non-sampling points, where the fitting patterns are patterns formed by sampling points within a set range from each non-sampling point;
a fitting mode determining module 13, configured to set a fitting mode of each non-sampling point according to fitting modes corresponding to the plurality of non-sampling points;
the fitting module 14 is configured to obtain a fitting result of each non-sampling point by using a fitting manner of each non-sampling point, where K space data of the plurality of sampling points and the fitting result of the non-sampling point form a fitting K space;
a reconstruction module 15 for reconstructing the fitted K-space to acquire a magnetic resonance image.
Optionally, the fitting mode determining module 12 is configured to determine, centering on each of the non-sampling points, a sampling point within a set range from each of the non-sampling points; the relative positions of the sampling points within a set range from each non-sampling point are combined to form a fitting pattern.
Optionally, the first fitting mode includes a K-space having a different location point than the second fitting mode.
Optionally, the number of sampling points included in the first fitting mode is different from the number of sampling points included in the second fitting mode.
Optionally, the fitting mode determining module is configured to classify fitting modes corresponding to a plurality of non-sampling points, where non-sampling points belonging to the same fitting mode are classified into one class; when the number of the non-sampling points in the same fitting mode is smaller than a set threshold value, setting the fitting mode of the corresponding non-sampling points to be K space fitting; and when the number of the non-sampling points in the same fitting mode is larger than a set threshold value, setting the fitting mode of the corresponding non-sampling points to be image domain fitting.
Optionally, the fitting mode setting module is configured to classify fitting modes corresponding to a plurality of non-sampling points, where non-sampling points belonging to the same fitting mode are classified into one class; calculating a first fitting time corresponding to the non-sampling points in the same fitting mode in a K space fitting mode; calculating a second fitting time corresponding to the non-sampling points in the same fitting mode in an image domain fitting mode; and selecting the fitting mode corresponding to the small one of the first fitting time and the second fitting time as the fitting mode of each non-sampling point.
Compared with the prior art, the technical scheme of the magnetic resonance image reconstruction device provided by the embodiment of the invention determines the fitting modes corresponding to the plurality of non-sampling points, sets the fitting mode of each non-sampling point according to the fitting modes corresponding to the plurality of non-sampling points, and acquires the fitting result of each non-sampling point by utilizing the fitting mode of each non-sampling point. By matching different fitting modes for different fitting modes, the fitting time of the non-sampling points of the different fitting modes is reduced, so that the fitting time of all the non-sampling points in the whole K space is reduced, the fitting speed of the non-sampling points in the K space is improved, and the technical effect of improving the reconstruction speed of the magnetic resonance image is achieved.
Example III
Fig. 5 is a schematic structural diagram of a magnetic resonance system according to a third embodiment of the present invention, as shown in fig. 5, which is
An embodiment of the present invention provides a magnetic resonance system, as shown in fig. 5, the system includes a scanning device 110, the scanning device 110 includes a radio frequency transmitting coil 111, a gradient coil 112, a radio frequency receiving coil 113, and a processor 120, the radio frequency transmitting coil 111 is used for transmitting a radio frequency pulse to a scanning part of a target object so as to excite nuclear spins of the scanning part; gradient coils 112 are used to apply slice selection gradient fields, phase encoding gradient fields, and frequency encoding gradient fields to the scan site to produce echo signals; the radio frequency receiving coil 113 is used for receiving echo signals to form magnetic resonance scanning data; the processor 120 is configured to obtain a K space to be processed, where the K space includes a plurality of sampling points and a plurality of non-sampling points; determining fitting modes corresponding to a plurality of non-sampling points, wherein the fitting modes are patterns formed by sampling points within a set range from each non-sampling point; setting a fitting mode of each non-sampling point according to fitting modes corresponding to the non-sampling points; obtaining a fitting result of each non-sampling point by using a fitting mode of each non-sampling point, wherein K space data of a plurality of sampling points and the fitting result of the non-sampling points form a fitting K space; the fit K-space is reconstructed to acquire a magnetic resonance image.
The K space to be processed is the K space currently used for image reconstruction, and the K space comprises a plurality of sampling points and at least two non-sampling points.
As shown in fig. 2, the non-sampled points can be obtained from the sampled points in the vicinity thereof by linear fitting, and the fitting formula is as follows:
S mn ＝∑ i,j C i,j S i,j (1)
wherein S is mn Representing the value of the non-sampling point obtained by fitting, wherein the non-sampling point corresponds to the m-th channel and the K space coordinate is n; c (C) i,j As the weight coefficient, S i,j Fitting values of sampling points within a range for non-sampling points, i representing the values used for fittingJ represents the K-space coordinates of the sample points for fitting, j+.n, and the sample points for fitting are near the K-space coordinates n. Weight coefficient C i,j The calibration data may be obtained by calculation, and the present embodiment is not described herein. In this embodiment, m and i are both positive integers, m=i for the case of single channel acquisition; for the case of multi-channel acquisition, i can take any channel value of 1, 2, 3, etc., and m is one of the multiple channels. In one embodiment, the coordinates of the K space may be any value from-127 to 128 in the phase encoding direction, -127. Ltoreq.j. Ltoreq.128, and-127. Ltoreq.n. Ltoreq.128.
It can be seen that when the number and distribution of sampling points around the non-sampling points are different, the weight coefficient matrix is also different, and the concept of fitting mode is introduced for this embodiment. The number and distribution of each sampling point is made to correspond to a fitting mode, namely, each weight coefficient matrix is made to correspond to a fitting mode. The non-sampled points (shaded points) in fig. 3A, 3B and 3C correspond to three different fitting patterns, respectively.
It will be appreciated that if the current sampling mode of the K-space is interlaced, the non-sampled points located in the middle portion of the K-space all correspond to the same fitting pattern, i.e., the fitting pattern in fig. 3A. If the current K-space data is sampled in a random manner, see fig. 3B and 3C, then the fit patterns corresponding to different non-sampled points of the K-space are typically different, or each fit pattern corresponds to only one or a small number of non-sampled points.
In one embodiment, if the distribution of sampling points within the fitting range of the non-sampling points is regular, the non-sampling points correspond to the first fitting pattern, see fig. 3A; if the distribution of the sampling points within the fitting range of the non-sampling points is irregular, the non-sampling points correspond to the second fitting pattern, see fig. 3B and 3C. It is understood that the position points of the K-space contained in the first fitting pattern are different from the position points of the K-space contained in the second fitting pattern. The first fitting pattern includes a different number of sampling points than the second fitting pattern.
According to the magnitude relation between the first fitting time and the second fitting time corresponding to each fitting mode, the fitting mode corresponding to each fitting mode is determined, so that the fitting time of all the non-sampling points corresponding to each fitting mode is the shortest.
The K-space fit corresponds to equation 1, and without loss of generality, can be represented as a convolution operation as follows:
wherein S is m Fitting result S for a plurality of non-sampled points mn Matrix formed, C' i Is C i,j Matrix obtained by coordinate inversion, S i Is S i,j The K-space data/matrix is composed,is a convolution operation.
According to the mathematical principle, the image domain fitting (image domain product fitting) corresponds to formula 2, and the convolution operation can be rewritten as:
S m ＝FFT(∑ i W i ·I i ) (3)
wherein W is i ＝IFFT(C' i )，I i ＝IFFT(S i ) FFT means Fourier transform, IFFT means Fourier inverse transform, and ". Cndot." means point multiplication operation.
It is verified that, for K-space data with uniform undersampling, such as fig. 3A, the sampled points and the non-sampled points are distributed in an interlaced manner along the phase encoding direction (the abscissa direction in the figure), a complete row is formed corresponding to a plurality of sampled points with the same phase encoding, a complete row is formed corresponding to a plurality of non-sampled points with the same phase encoding, and the operation speed of formula 3 is faster than that of formula 2, but in the case that the data matrix contained in the K-space is smaller or the number of non-sampled points to be fitted is small, or in the case of non-uniform sampling, as shown in fig. 3B and 3C, the operation speed of formula 2 is faster.
It can be seen that for any non-sampled point, the magnitude relationship between its corresponding two fitting times is related to its corresponding number of sampled points or whether its corresponding distribution of sampled points is regular. For a plurality of non-sampling points with irregular sampling point distribution, the embodiment classifies fitting modes corresponding to the non-sampling points, and the non-sampling points belonging to the same fitting mode are classified into one type; and calculating a first fitting time corresponding to the non-sampling points in the same fitting mode in a K space fitting mode and a second fitting time corresponding to the non-sampling points in the same fitting mode in an image domain mode, and then selecting the fitting mode corresponding to the small one of the first fitting time and the second fitting time as the fitting mode of each non-sampling point.
If the sampling point distribution rule corresponding to the non-sampling point is not adopted, the corresponding first fitting mode is directly judged, and the fitting mode is set as the image domain fitting mode.
If the distribution of the sampling points corresponding to the fitting mode is irregular, the fitting time is related to the number of the sampling points corresponding to the fitting mode. Therefore, after the fitting mode of each non-sampling point is determined, the fitting mode corresponding to each non-sampling point needs to be determined according to the magnitude relation between the number of sampling points corresponding to the fitting mode and the preset threshold. Specifically, if the number of non-sampled points in the same fitting mode is smaller than a set threshold, it is known that the fitting time of the corresponding K-space fitting mode is smaller than the fitting time of the corresponding image domain, and then the fitting mode of the non-sampled points in the fitting mode is set as a K-space fitting mode; if the number of the non-sampling points in the same fitting mode is larger than a set threshold, the fitting time of the corresponding K space fitting mode is larger than or equal to the fitting time of the corresponding image domain, and at the moment, the fitting mode of the non-sampling points in the fitting mode is set as the image domain fitting mode.
Wherein the magnitude of the set threshold is related to the image processing parameters of the processor. If the configuration of the image processing parameters of the processor is higher, the image processing speed is higher, and the fitting mode with a little more fitting points is fitted by adopting a K space fitting mode, the set threshold will be higher if less effort is required, otherwise, if the configuration of the performance parameters of the processor is lower, the image processing speed is slower, and the fitting speed is rapidly reduced when the fitting mode with a little more fitting points is fitted by adopting a K space fitting mode, so the set threshold will be lower.
Fitting each non-sampling point by using a fitting mode of each non-sampling point so as to obtain a fitting result of each non-sampling point. Fitting results of all the non-sampling points and K space data of a plurality of sampling points form a fitting space.
In some embodiments, after determining the fitting manner corresponding to the non-sampling points of each fitting mode, fitting all the non-sampling points corresponding to each fitting mode by adopting the fitting manner corresponding to each fitting mode, so as to obtain the fitting result of all the non-sampling points corresponding to the same fitting mode at one time. After the fitting operation is performed on all the non-sampled points corresponding to all the fitting modes, the fitting result of all the non-sampled points in the K space can be obtained.
In one embodiment, all fitting modes are traversed, the fitting point number of the current fitting mode is determined, the fitting mode corresponding to the current fitting mode is determined according to the size relation between the fitting point number and a set threshold value, and fitting is carried out on all non-sampled points corresponding to the current fitting mode by adopting the fitting mode, so that the fitting result of all non-sampled points corresponding to the current fitting mode is obtained.
For the image domain fitting mode, the direct fitting result is image domain data, the fitting result is required to be transformed into K space to obtain a transformation result, fitting data of one or more non-sampling points corresponding to the corresponding fitting mode are extracted from the transformation result, and the extracted fitting data are respectively filled into the corresponding non-sampling points of the K space to update the fitting result corresponding to the current fitting mode.
After the fitted K space is obtained, the fitted K space is subjected to image reconstruction to obtain a magnetic resonance image.
Compared with the prior art, the technical scheme of the magnetic resonance image reconstruction method provided by the embodiment of the invention determines the fitting modes corresponding to the plurality of non-sampling points, sets the fitting mode of each non-sampling point according to the fitting modes corresponding to the plurality of non-sampling points, and acquires the fitting result of each non-sampling point by utilizing the fitting mode of each non-sampling point. By matching different fitting modes for different fitting modes, the fitting time of the non-sampling points of the different fitting modes is reduced, so that the fitting time of all the non-sampling points in the whole K space is reduced, the fitting speed of the non-sampling points in the K space is improved, and the technical effect of improving the reconstruction speed of the magnetic resonance image is achieved.
As shown in fig. 5, the magnetic resonance imaging system 100 further includes a controller 130, an input device 140 and an output device 150, wherein the controller 130 may include one or a combination of several of a central controller (Central Processing Unit, CPU), application-specific integrated circuit (ASIC), special instruction controller (Application Specific Instruction Set Processor, ASIP), graphics processing unit (Graphics Processing Unit, GPU), physical controller (Physics Processing Unit, PPU), digital signal controller (Digital Processing Processor, DSP), field-programmable gate array (Field-Programmable Gate Array, FPGA), ARM controller, etc.
An output device 150, such as a display, may display the magnetic resonance image of the region of interest. Further, the output device 150 may also display the height, weight, age, imaging location, and operating status of the scanning device 110 of the subject. The type of the output device 150 may be one or a combination of several of a Cathode Ray Tube (CRT) output device, a liquid crystal output device (LCD), an organic light emitting output device (OLED), a plasma output device, etc. In an embodiment, the display is configured to display a recommended value of a protocol parameter in response to scanning the protocol parameter so that a target object has a PNS overrun slice to be scanned in a magnetic resonance data acquisition process, the recommended value including: one or more of gradient rotation angle recommendation, gradient recommendation mode, bandwidth recommendation, readout resolution recommendation, phase resolution recommendation.
The magnetic resonance imaging system 100 may be connected to a local area network (Local Area Network, LAN), wide area network (Wide Area Network, WAN), public network, private network, proprietary network, public switched telephone network (Public Switched Telephone Network, PSTN), the internet, wireless network, virtual network, or any combination thereof.
The scanning device 110 comprises an MR signal acquisition module, an MR control module and an MR data storage module. The MR signal acquisition module comprises a magnet unit and a radio frequency unit. The magnet unit mainly includes a main magnet generating a B0 main magnetic field and a gradient assembly generating a gradient. The main magnet contained in the magnet unit may be a permanent magnet or a superconducting magnet, the gradient assembly mainly comprises gradient current Amplifiers (AMPs), gradient coils, and the gradient assembly may further comprise three independent channels Gx, gy, gz, each gradient amplifier exciting a corresponding one of the gradient coils in the gradient coil set to generate gradient fields for generating corresponding spatially encoded signals for spatially localization of the magnetic resonance signals. The radio frequency unit mainly comprises a radio frequency transmitting coil and a radio frequency receiving coil, wherein the radio frequency transmitting coil is used for transmitting radio frequency pulse signals to a person to be detected or a human body, the radio frequency receiving coil is used for receiving magnetic resonance signals acquired from the human body, and the radio frequency coils forming the radio frequency unit can be divided into a body coil and a local coil according to different functions. In one embodiment, the type of body coil or local coil may be a birdcage coil, a solenoid coil, a saddle coil, a helmholtz coil, an array coil, a loop coil, or the like. In one particular embodiment, the local coils are provided as array coils, and the array coils may be provided in a 4-channel mode, an 8-channel mode, or a 16-channel mode. The magnet unit and the radio frequency unit may constitute an open low field magnetic resonance device or a closed superconducting magnetic resonance device.
The MR control module may monitor an MR signal acquisition module, an MR data processing module, comprising a magnet unit and a radio frequency unit. Specifically, the MR control module may receive the information or pulse parameters sent by the MR signal acquisition module; in addition, the MR control module can also control the processing procedure of the MR data processing module. In one embodiment, the MR control module is further connected to a pulse sequence generator, a gradient waveform generator, a transmitter, a receiver, etc., and controls the magnetic field module to execute a corresponding scanning sequence after receiving instructions from the console.
Illustratively, the specific process of generating MR data by the scanning device 110 in the embodiments of the present invention includes: the main magnet generates a B0 main magnetic field, and atomic nuclei in the subject generate precession frequency under the action of the main magnetic field, wherein the precession frequency is in direct proportion to the intensity of the main magnetic field; the MR control module stores and transmits an instruction of a scan sequence to be executed, the pulse sequence generator controls the gradient waveform generator and the transmitter according to the scan sequence instruction, the gradient waveform generator outputs gradient pulse signals with preset time sequences and waveforms, the signals pass through Gx, gy and Gz gradient current amplifiers, and then pass through three independent channels Gx, gy and Gz in the gradient assembly, and each gradient amplifier excites a corresponding gradient coil in the gradient coil group to generate a gradient field for generating corresponding spatial coding signals so as to spatially locate magnetic resonance signals; the pulse sequence generator also executes a scanning sequence, outputs data including timing, intensity, shape and the like of radio frequency pulses transmitted by radio frequency, timing of radio frequency reception and length of a data acquisition window to the transmitter, simultaneously the transmitter transmits corresponding radio frequency pulses to a body transmitting coil in the radio frequency unit to generate a B1 field, signals emitted by excited atomic nuclei in a patient body under the action of the B1 field are perceived by a receiving coil in the radio frequency unit, and then transmitted to an MR data processing module through a transmitting/receiving switch, and is subjected to digital processing such as amplification, demodulation, filtering, AD conversion and the like, and then transmitted to an MR data storage module. The scan is completed when the MR data storage module acquires a set of raw k-space data. The raw k-space data is rearranged into separate k-space data sets corresponding to each image to be reconstructed, each k-space data set is input to an array controller, and the image reconstruction is performed to combine the magnetic resonance signals to form a set of image data.
Example IV
The embodiments of the present invention also provide a storage medium containing computer executable instructions which, when executed by a computer processor, are for performing a magnetic resonance image reconstruction method comprising:
acquiring all non-sampling points of a K space, and determining at least one fitting mode corresponding to all non-sampling points in the K space, wherein the K space is one of at least one target block contained in current K space data;
determining the minimum fitting time of at least two selectable fitting times corresponding to each fitting mode, and taking the fitting mode corresponding to the minimum fitting time as the fitting mode of each fitting mode;
fitting all the non-sampled points corresponding to each fitting mode by adopting a fitting mode corresponding to each fitting mode so as to obtain fitting results of all the non-sampled points in the K space;
repeating the steps until fitting results of all non-sampling points in the current K space data are obtained, so as to update the current K space data, and carrying out image reconstruction on the updated K space data to obtain a magnetic resonance image.
Of course, the storage medium containing the computer executable instructions provided in the embodiments of the present invention is not limited to the method operations described above, and may also perform the related operations in the magnetic resonance image reconstruction method provided in any embodiment of the present invention.
From the above description of embodiments, it will be clear to a person skilled in the art that the present invention may be implemented by means of software and necessary general purpose hardware, but of course also by means of hardware, although in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, where the instructions include a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the magnetic resonance image reconstruction method according to the embodiments of the present invention.
It should be noted that, in the embodiment of the magnetic resonance image reconstruction device, each unit and module included are only divided according to the functional logic, but are not limited to the above-mentioned division, so long as the corresponding functions can be implemented; in addition, the specific names of the functional units are also only for distinguishing from each other, and are not used to limit the protection scope of the present invention.
Note that the above is only a preferred embodiment of the present invention and the technical principle applied. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, while the invention has been described in connection with the above embodiments, the invention is not limited to the embodiments, but may be embodied in many other equivalent forms without departing from the spirit or scope of the invention, which is set forth in the following claims.
Claims (9)
1. A method of magnetic resonance image reconstruction, comprising:
acquiring a K space to be processed, wherein the K space comprises a plurality of sampling points and a plurality of non-sampling points;
determining a fitting mode corresponding to a plurality of non-sampling points, wherein the fitting mode is a pattern formed by sampling points within a set range from each non-sampling point; the fitting modes are patterns formed by sampling points within a set range from each non-sampling point, the patterns are patterns formed by the number and the distribution of sampling points around the non-sampling points, and when the number and the distribution of the sampling points around the non-sampling points are different, the fitting modes are different, so that the number and the distribution of the sampling points around each non-sampling point correspond to one fitting mode;
Setting a fitting mode of each non-sampling point according to fitting modes corresponding to the non-sampling points;
obtaining a fitting result of each non-sampling point by using a fitting mode of each non-sampling point, wherein K space data of the plurality of sampling points and the fitting result of the non-sampling points form a fitting K space;
reconstructing the fit K-space to obtain a magnetic resonance image;
setting a fitting mode of each non-sampling point according to fitting modes corresponding to the non-sampling points, wherein the fitting mode comprises the following steps:
classifying fitting modes corresponding to the plurality of non-sampling points, and dividing the non-sampling points belonging to the same fitting mode into one class;
calculating a first fitting time corresponding to the non-sampling points in the same fitting mode in a K space fitting mode;
calculating a second fitting time corresponding to the non-sampling points in the same fitting mode in an image domain fitting mode;
and selecting the fitting mode corresponding to the small one of the first fitting time and the second fitting time as the fitting mode of each non-sampling point.
2. The method of claim 1, wherein setting a fitting manner for each of the plurality of non-sampled points according to the fitting pattern for the plurality of non-sampled points comprises:
and selecting one reconstruction method for each non-sampling point from at least two different reconstruction methods according to the fitting modes corresponding to the non-sampling points, wherein at least two different reconstruction methods are used in the K space reconstruction process.
3. The method of claim 1, wherein the fitting pattern for the plurality of non-sampled points is determined by:
taking each non-sampling point as a center, and determining a sampling point within a set range from each non-sampling point;
the combination of the relative positions of the sampling points within the set range from each non-sampling point forms a fitting pattern.
4. The method of claim 1, wherein the fitting pattern for the plurality of non-sampled points comprises a first fitting pattern and a second fitting pattern, the first fitting pattern corresponding to a regular pattern and the second fitting pattern corresponding to an irregular pattern.
5. The method of claim 4, wherein the first fit pattern comprises a different location point of K-space than the second fit pattern.
6. The method of claim 1, wherein setting a fitting manner for each of the non-sampled points according to the fitting patterns corresponding to the plurality of non-sampled points comprises:
counting the non-sampled points with the same fitting mode;
when the number of the non-sampling points in the same fitting mode is smaller than a set threshold value, setting the fitting mode of the corresponding non-sampling points to be K space fitting;
And when the number of the non-sampling points in the same fitting mode is larger than a set threshold value, setting the fitting mode of the corresponding non-sampling points to be image domain fitting.
7. A magnetic resonance image reconstruction apparatus, comprising:
the acquisition module is used for acquiring a K space to be processed, wherein the K space comprises a plurality of sampling points and non-sampling points;
the fitting mode determining module is used for determining fitting modes corresponding to a plurality of non-sampling points, wherein the fitting modes are patterns formed by sampling points within a set range from each non-sampling point; the fitting modes are patterns formed by sampling points within a set range from each non-sampling point, the patterns are patterns formed by the number and the distribution of sampling points around the non-sampling points, and when the number and the distribution of the sampling points around the non-sampling points are different, the fitting modes are different, so that the number and the distribution of the sampling points around each non-sampling point correspond to one fitting mode;
the fitting mode determining module is used for setting the fitting mode of each non-sampling point according to the fitting modes corresponding to the non-sampling points;
the fitting mode setting module is specifically used for classifying fitting modes corresponding to a plurality of non-sampling points, and the non-sampling points belonging to the same fitting mode are classified into one type; calculating a first fitting time corresponding to the non-sampling points in the same fitting mode in a K space fitting mode; calculating a second fitting time corresponding to the non-sampling points in the same fitting mode in an image domain fitting mode; selecting the fitting mode corresponding to the small one of the first fitting time and the second fitting time as the fitting mode of each non-sampling point
The fitting module is used for obtaining a fitting result of each non-sampling point by using a fitting mode of each non-sampling point, and K space data of the plurality of sampling points and the fitting result of the non-sampling points form a fitting K space;
and the reconstruction module is used for reconstructing the fitting K space so as to acquire a magnetic resonance image.
8. A magnetic resonance system, comprising:
the radio frequency transmitting coil is used for transmitting radio frequency pulses to a scanning part of a target object so as to excite nuclear spins of the scanning part;
gradient coils for applying slice selection gradient fields, phase encoding gradient fields, and frequency encoding gradient fields to the scan site to generate echo signals;
a radio frequency receiving coil for receiving the echo signals to form magnetic resonance scan data;
the processor is used for acquiring a K space to be processed, wherein the K space comprises a plurality of sampling points and non-sampling points; determining a fitting mode corresponding to a plurality of non-sampling points, wherein the fitting mode is a pattern formed by sampling points within a set range from each non-sampling point; the fitting modes are patterns formed by sampling points within a set range from each non-sampling point, the patterns are patterns formed by the number and the distribution of sampling points around the non-sampling points, and when the number and the distribution of the sampling points around the non-sampling points are different, the fitting modes are different, so that the number and the distribution of the sampling points around each non-sampling point correspond to one fitting mode; setting a fitting mode of each non-sampling point according to fitting modes corresponding to the non-sampling points; obtaining a fitting result of each non-sampling point by using a fitting mode of each non-sampling point, wherein K space data of the plurality of sampling points and the fitting result of the non-sampling points form a fitting K space; reconstructing the fitting K space to obtain a magnetic resonance image, and setting a fitting mode of each non-sampling point according to fitting modes corresponding to the non-sampling points, including: classifying fitting modes corresponding to the plurality of non-sampling points, and dividing the non-sampling points belonging to the same fitting mode into one class; calculating a first fitting time corresponding to the non-sampling points in the same fitting mode in a K space fitting mode; calculating a second fitting time corresponding to the non-sampling points in the same fitting mode in an image domain fitting mode; and selecting the fitting mode corresponding to the small one of the first fitting time and the second fitting time as the fitting mode of each non-sampling point.
9. A storage medium containing computer executable instructions which, when executed by a computer processor, are for performing the magnetic resonance image reconstruction method as claimed in any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011051798.4A CN114325524B (en) | 2020-09-29 | 2020-09-29 | Magnetic resonance image reconstruction method, device, system and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011051798.4A CN114325524B (en) | 2020-09-29 | 2020-09-29 | Magnetic resonance image reconstruction method, device, system and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114325524A CN114325524A (en) | 2022-04-12 |
| CN114325524B true CN114325524B (en) | 2023-09-01 |
Family
ID=81010839
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011051798.4A Active CN114325524B (en) | 2020-09-29 | 2020-09-29 | Magnetic resonance image reconstruction method, device, system and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114325524B (en) |
Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2002048731A2 (en) * | 2000-12-11 | 2002-06-20 | THE GOVERNEMENT OF THE UNITED STATES OF AMERICA, as represented by THE SECRETARY, DEPARTMENT OF HEALTH AND HUMAN SERVICES | Accelerated magnetic resonance imaging |
| US7215731B1 (en) * | 2005-11-30 | 2007-05-08 | General Electric Company | Fast backprojection/reprojection with hexagonal segmentation of image |
| CN101635446A (en) * | 2009-06-08 | 2010-01-27 | 国电南瑞科技股份有限公司 | Method for processing sample data based on waveform fitting and secondary sampling |
| JP2011128784A (en) * | 2009-12-16 | 2011-06-30 | Mitsubishi Electric Corp | Rpc calculation device |
| CN103064046A (en) * | 2012-12-25 | 2013-04-24 | 深圳先进技术研究院 | Image processing method based on sparse sampling magnetic resonance imaging |
| 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 |
| CN103472419A (en) * | 2013-08-30 | 2013-12-25 | 深圳先进技术研究院 | Magnetic-resonance fast imaging method and system thereof |
| CN103679654A (en) * | 2013-11-29 | 2014-03-26 | 深圳先进技术研究院 | Magnetic resonance imaging method and system |
| CN104635188A (en) * | 2013-11-12 | 2015-05-20 | 上海联影医疗科技有限公司 | K-space reconstruction method and magnetic resonance imaging method |
| WO2015119328A1 (en) * | 2014-02-05 | 2015-08-13 | Gachon University Of Industry-Academic Cooperation Foundation | Magnetic resonance imaging method using t2* mapping based on echo planar imaging |
| CN106803275A (en) * | 2017-02-20 | 2017-06-06 | 苏州中科广视文化科技有限公司 | Estimated based on camera pose and the 2D panoramic videos of spatial sampling are generated |
| CN107576925A (en) * | 2017-08-07 | 2018-01-12 | 上海东软医疗科技有限公司 | The more contrast image rebuilding methods of magnetic resonance and device |
| CN107656224A (en) * | 2017-09-30 | 2018-02-02 | 上海联影医疗科技有限公司 | A kind of MR imaging method, apparatus and system |
| CN108896943A (en) * | 2018-05-10 | 2018-11-27 | 上海东软医疗科技有限公司 | A kind of magnetic resonance quantitative imaging method and device |
| CN109239631A (en) * | 2018-08-03 | 2019-01-18 | 上海东软医疗科技有限公司 | A kind of MR imaging method and device |
| CN109350098A (en) * | 2018-08-27 | 2019-02-19 | 苏州瑞派宁科技有限公司 | Determination method, method for reconstructing and the device of the fit approach of signal |
| CN110161442A (en) * | 2018-02-12 | 2019-08-23 | 深圳先进技术研究院 | Magnetic resonance parameters imaging method, device, medical supply and storage medium |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1799498B (en) * | 2004-12-31 | 2010-04-28 | 西门子（中国）有限公司 | Quick generalized partially parallel acquisition(GRAPA) image reconstruction algorithm for magnetic resonance imaging(MRI) |
| US9224196B2 (en) * | 2014-03-12 | 2015-12-29 | ClearMark Systems, LLC | System and method for authentication |
| US10775464B2 (en) * | 2017-01-31 | 2020-09-15 | Regents Of The University Of Minnesota | System and method for dynamic, cardiac phase-resolved quantitative longitudinal relaxation parameter mapping |
| US10726552B2 (en) * | 2017-03-20 | 2020-07-28 | The General Hospital Corporation | Quantification of magnetic resonance data by adaptive fitting of downsampled images |
-
2020
- 2020-09-29 CN CN202011051798.4A patent/CN114325524B/en active Active
Patent Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2002048731A2 (en) * | 2000-12-11 | 2002-06-20 | THE GOVERNEMENT OF THE UNITED STATES OF AMERICA, as represented by THE SECRETARY, DEPARTMENT OF HEALTH AND HUMAN SERVICES | Accelerated magnetic resonance imaging |
| US7215731B1 (en) * | 2005-11-30 | 2007-05-08 | General Electric Company | Fast backprojection/reprojection with hexagonal segmentation of image |
| CN101635446A (en) * | 2009-06-08 | 2010-01-27 | 国电南瑞科技股份有限公司 | Method for processing sample data based on waveform fitting and secondary sampling |
| JP2011128784A (en) * | 2009-12-16 | 2011-06-30 | Mitsubishi Electric Corp | Rpc calculation device |
| 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 |
| CN103064046A (en) * | 2012-12-25 | 2013-04-24 | 深圳先进技术研究院 | Image processing method based on sparse sampling magnetic resonance imaging |
| CN103472419A (en) * | 2013-08-30 | 2013-12-25 | 深圳先进技术研究院 | Magnetic-resonance fast imaging method and system thereof |
| CN104635188A (en) * | 2013-11-12 | 2015-05-20 | 上海联影医疗科技有限公司 | K-space reconstruction method and magnetic resonance imaging method |
| CN103679654A (en) * | 2013-11-29 | 2014-03-26 | 深圳先进技术研究院 | Magnetic resonance imaging method and system |
| WO2015119328A1 (en) * | 2014-02-05 | 2015-08-13 | Gachon University Of Industry-Academic Cooperation Foundation | Magnetic resonance imaging method using t2* mapping based on echo planar imaging |
| CN106803275A (en) * | 2017-02-20 | 2017-06-06 | 苏州中科广视文化科技有限公司 | Estimated based on camera pose and the 2D panoramic videos of spatial sampling are generated |
| CN107576925A (en) * | 2017-08-07 | 2018-01-12 | 上海东软医疗科技有限公司 | The more contrast image rebuilding methods of magnetic resonance and device |
| CN107656224A (en) * | 2017-09-30 | 2018-02-02 | 上海联影医疗科技有限公司 | A kind of MR imaging method, apparatus and system |
| CN110161442A (en) * | 2018-02-12 | 2019-08-23 | 深圳先进技术研究院 | Magnetic resonance parameters imaging method, device, medical supply and storage medium |
| CN108896943A (en) * | 2018-05-10 | 2018-11-27 | 上海东软医疗科技有限公司 | A kind of magnetic resonance quantitative imaging method and device |
| CN109239631A (en) * | 2018-08-03 | 2019-01-18 | 上海东软医疗科技有限公司 | A kind of MR imaging method and device |
| CN109350098A (en) * | 2018-08-27 | 2019-02-19 | 苏州瑞派宁科技有限公司 | Determination method, method for reconstructing and the device of the fit approach of signal |
Non-Patent Citations (1)
| Title |
|---|
| 葛永新等.联合局部和全局稀疏表示的磁共振图像重建方法.《重庆大学学报》.2017,第40卷(第1期),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114325524A (en) | 2022-04-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10185012B2 (en) | System and method for magnetic resonance imaging | |
| JP6513398B2 (en) | MR image reconstruction using prior information constrained regularization | |
| US7015696B2 (en) | Magnetic resonance imaging apparatus and magnetic resonance imaging method | |
| EP1391746A2 (en) | Parallel magnetic resonance imaging using navigator echos | |
| CN107209238B (en) | Parallel multi-slice MR imaging with suppression of banding artifacts | |
| US11067653B2 (en) | Magnetic resonance imaging method including acquiring echo signals from each k-space slice with different relaxation time weightings after echo signals from different k-space splices are acquired | |
| JP6356809B2 (en) | Zero echo time MR imaging with water / fat separation | |
| CN109188326B (en) | Magnetic resonance imaging method and magnetic resonance system | |
| WO2012077543A1 (en) | Magnetic resonance imaging device and contrast-enhanced image acquisition method | |
| EP3295194B1 (en) | Parallel mr imaging with rf coil sensitivity mapping | |
| CN110664378B (en) | Magnetic resonance imaging method, device, system and storage medium | |
| CN108459289A (en) | A kind of multiple excitation Diffusion weighted MR imaging method based on data consistency | |
| US20170350954A1 (en) | Spin echo mr imaging | |
| JP2017529960A (en) | Propeller MR imaging with artifact suppression | |
| CN110095742A (en) | A kind of echo planar imaging neural network based and device | |
| CN110215209B (en) | Magnetic resonance imaging method and magnetic resonance imaging system | |
| EP2856196A2 (en) | System and method for magnetic resonance imaging using highly accelerated projection imaging | |
| CN109975343B (en) | Magnetic resonance imaging method, system and storage medium | |
| US20090206835A1 (en) | Magnetic Resonance Imaging Apparatus and Method | |
| US11959986B2 (en) | MR imaging with spiral acquisition | |
| EP3673282B1 (en) | Data-driven correction of phase depending artefacts in a magnetic resonance imaging system | |
| CN114325524B (en) | Magnetic resonance image reconstruction method, device, system and storage medium | |
| CN113945877B (en) | Magnetic resonance black blood imaging method and system | |
| JP2020522344A (en) | Parallel multi-slice MR imaging | |
| CN113534032B (en) | Magnetic resonance imaging method and system |
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 |