CN112927313A - Magnetic resonance image reconstruction method, magnetic resonance image reconstruction device, computer equipment and readable storage medium - Google Patents
Magnetic resonance image reconstruction method, magnetic resonance image reconstruction device, computer equipment and readable storage medium Download PDFInfo
- Publication number
- CN112927313A CN112927313A CN201911235888.6A CN201911235888A CN112927313A CN 112927313 A CN112927313 A CN 112927313A CN 201911235888 A CN201911235888 A CN 201911235888A CN 112927313 A CN112927313 A CN 112927313A
- Authority
- CN
- China
- Prior art keywords
- reconstruction
- image
- constraint
- initial image
- data set
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/003—Reconstruction from projections, e.g. tomography
- G06T11/006—Inverse problem, transformation from projection-space into object-space, e.g. transform methods, back-projection, algebraic methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2211/00—Image generation
- G06T2211/40—Computed tomography
- G06T2211/424—Iterative
Abstract
The application relates to a magnetic resonance image reconstruction method, a magnetic resonance image reconstruction device, a computer device and a readable storage medium. The magnetic resonance image reconstruction method comprises the following steps: acquiring a K space data set corresponding to a magnetic resonance signal of an object, wherein the K space data set is acquired in an undersampling mode; performing primary reconstruction on the K space data set to obtain an initial image; based on a set constraint, performing iterative reconstruction on the K-space data set and the initial image to obtain a target image of the object, wherein the reconstruction of the K-space data set corresponds to the whole of the initial image, and the reconstruction of the initial image corresponds to the local of the initial image. The method provided by the application can improve the efficiency of magnetic resonance image reconstruction.
Description
Technical Field
The present application relates to the field of magnetic resonance imaging technologies, and in particular, to a magnetic resonance image reconstruction method and apparatus, a computer device, and a readable storage medium.
Background
Magnetic Resonance Imaging (MRI) is an imaging technique that utilizes the signals generated by the resonance of atomic nuclei within a strong magnetic field for image reconstruction. The principle of magnetic resonance imaging is that a radio frequency pulse is used for exciting an atomic nucleus which contains a non-zero spin and is arranged in a magnetic field, the atomic nucleus is relaxed after the radio frequency pulse is stopped, an induction coil is used for collecting signals in the relaxation process, and a mathematical image is reconstructed according to a certain mathematical method.
Magnetic resonance imaging is widely applied, but the scanning speed of magnetic resonance imaging is always a big problem restricting the magnetic resonance imaging. Magnetic resonance accelerated imaging is therefore becoming of particular importance. The mainstream magnetic resonance accelerated imaging algorithm mainly comprises a partial Fourier acceleration algorithm, a parallel acquisition acceleration algorithm, a compressed sensing acceleration algorithm and an artificial intelligence acceleration algorithm. The acceleration algorithms pay attention to partial prior knowledge, and the undersampled data is recovered by using reasonable algorithm constraint so as to achieve the acceleration effect. In the prior art, some algorithms fuse several algorithms, but the algorithms still have the problem of low imaging efficiency.
Disclosure of Invention
In view of the above, it is necessary to provide a magnetic resonance image reconstruction method, apparatus, computer device and readable storage medium.
In a first aspect, an embodiment of the present application provides a magnetic resonance image reconstruction method, including:
acquiring a K space data set corresponding to a magnetic resonance signal of a subject;
performing preliminary reconstruction on the K space data set to obtain an initial image, wherein the K space data set is obtained in an undersampling mode;
based on a set constraint, performing iterative reconstruction on the K-space data set and the initial image to obtain a target image of the object, wherein the reconstruction of the K-space data set corresponds to the whole of the initial image, and the reconstruction of the initial image corresponds to the local of the initial image.
Optionally, the performing iterative reconstruction on the K-space data set and the initial image based on the set constraint to obtain a target image of the object includes:
establishing an image reconstruction model based on the set constraints, wherein the set constraints comprise at least two types, the reconstruction region of one type of constraint comprises the global part of the initial image, and the reconstruction region of the other type of constraint comprises only the local part of the initial image;
and performing iterative reconstruction on the K space data set and the initial image according to the image reconstruction model to obtain the target image.
Optionally, the establishing an image reconstruction model based on the set constraint includes:
setting a weight parameter of each set constraint;
and establishing the image reconstruction model based on the weight parameters of the set constraints.
Optionally, the image reconstruction model is:
min∑ρifi,j,k(x)
where x represents the image pixel value of the target region, fi,j,k(x) Representing the i-th constraint, j representing the space in which the i-th constraint is performed, k representing the reconstruction region of the i-th constraint, piA weight parameter representing said ith constraint.
Optionally, the set constraint includes at least two of a partial fourier constraint, a merging constraint, a compressive sensing constraint and an artificial intelligence constraint.
Optionally, the number of the target images is at least two, and the at least two reconstruction regions with set constraints respectively correspond to different parts of the initial image.
Optionally, the image reconstruction model is:
wherein x is data of the initial image, y is the K space data set, T is global reconstruction of the initial image, P is local reconstruction of the initial image, F is Fourier transformation constraint, G is K space parallel acquisition constraint, C is K space conjugate symmetry constraint, xe is data of the initial image after phase calibration, CS is compressive sensing constraint, AI is artificial intelligence constraint, and lambda is a linear transformation constraintiAs a weighting factor, λiSi is the ith seedA confidence factor of the beam.
In a second aspect, a magnetic resonance image reconstruction apparatus is provided, including:
the data acquisition module is used for acquiring a K space data set corresponding to a magnetic resonance signal of an object, wherein the K space data set is acquired in an undersampling mode;
the preliminary reconstruction module is used for carrying out preliminary reconstruction on the K space data set to obtain an initial image;
an iterative reconstruction module, configured to perform iterative reconstruction on the K-space data set and the initial image based on a set constraint to obtain a target image of the object, where the reconstruction of the K-space data set corresponds to a global state of the initial image, and the reconstruction of the initial image corresponds to a local state of the initial image.
In a third aspect, a computer device is provided, comprising a memory storing a computer program and a processor implementing the steps of the method as described above when the processor executes the computer program.
In a fourth aspect, a computer-readable storage medium is provided, having stored thereon a computer program which, when being executed by a processor, carries out the steps of the method as described above.
The magnetic resonance image reconstruction method, the magnetic resonance image reconstruction device, the computer device, and the readable storage medium provided by the embodiments obtain an initial image by performing preliminary reconstruction on the K-space data set, and perform iterative reconstruction on the K-space data set and the initial image by using different constraint algorithms based on set constraints to obtain a target image of an object. The reconstruction of the K space data set corresponds to the whole of the initial image, and the reconstruction of the initial image corresponds to the local of the initial image, so that different constraints can concern different areas when the constraint algorithm is fused and reconstructed, and the calculation efficiency is improved; the reconstruction of the initial image only corresponds to the local part of the initial image, and the overall reconstruction is not needed, so that the calculation efficiency is further improved, and the magnetic resonance imaging efficiency is improved.
Drawings
Figure 1 is a schematic flow chart of a magnetic resonance image reconstruction in an embodiment;
FIG. 2 is a schematic illustration of an initial image and a target image in one embodiment;
figure 3 is a schematic flow chart of a magnetic resonance image reconstruction in an embodiment;
figure 4 is a schematic flow chart of a magnetic resonance image reconstruction in an embodiment;
fig. 5 is a schematic structural diagram of a magnetic resonance image reconstruction apparatus according to an embodiment;
FIG. 6 is a diagram illustrating an internal structure of a computer device according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The magnetic resonance image reconstruction method provided by the embodiment of the application can be applied to a magnetic resonance imaging device and is used for reconstructing an image according to data corresponding to a magnetic resonance signal after the magnetic resonance signal of an object is acquired. The magnetic resonance image reconstruction method provided by the embodiment of the application can be particularly applied to computer equipment, and the computer equipment can be but is not limited to various personal computers, laptops, smartphones, tablets and portable wearable equipment. The computer device comprises a memory capable of storing data and a computer program and a processor capable of executing the computer program to implement the magnetic resonance image reconstruction method provided by the embodiments of the present application. The magnetic resonance image reconstruction method is further described in detail with reference to specific embodiments.
Referring to fig. 1, an embodiment of the present application provides a magnetic resonance image reconstruction method, including:
s10, acquiring a K space data set corresponding to the magnetic resonance signal of the object, wherein the K space data set is acquired in an undersampling mode;
s20, carrying out primary reconstruction on the K space data set to obtain an initial image;
s30, based on the set constraint, performing iterative reconstruction on the K-space dataset and the initial image to obtain a target image of the object, the reconstruction of the K-space dataset corresponding to the global of the initial image, and the reconstruction of the initial image corresponding to the local of the initial image.
The object refers to a human or animal body that is subject to magnetic resonance detection and scanning. The magnetic resonance imaging device scans an object, acquires magnetic resonance signals, and acquires a plurality of groups of K space data to obtain a K space data set. And the K space data set acquired by the magnetic resonance imaging device is transmitted to the computer equipment, and the computer equipment carries out primary image reconstruction based on the K space data set to obtain a primary image of the object. The preliminary image reconstruction method can be selected according to actual conditions, and the method is not limited in any way in the application.
The target image refers to an image of a region of interest or a region of interest, for example, a head image. After the preliminary image is obtained, the computer device further performs further iterative reconstruction on the K space data set and the initial image based on the set constraint so as to obtain a target image. The set constraints of the iterative reconstruction can be selected according to actual requirements. The set constraints are different, and the corresponding reconstruction algorithms can be different; the set constraints are different, and the space where the constraints are performed can be different, for example, the constraints can be performed in a K space, an image domain, or a mixed domain (K space and/or image domain); the region in which the reconstruction is performed may be different depending on the setting constraints, and for example, the reconstruction may be performed in the entire region or only in the region of interest. In this embodiment, iterative reconstruction is performed on the K-space data set, and the reconstructed region is a whole region, that is, the whole of the initial image. And performing iterative reconstruction on the initial image, and constraining the initial image in an image domain, namely the local part of the initial image.
Referring to fig. 2, fig. 2 is an image reconstructed from an oversampled K-space data set. In fig. 2, the area of the whole image is a global area, and the corresponding image is the initial image. The partial area of the image is shown in the white frame of the figure, corresponding to the target image. When iterative reconstruction is performed on a K space data set and an initial image, under the condition that an algorithm frame corresponding to constraint is set, firstly determining a region needing attention of each algorithm, such as a Parallel Acquisition Technology (PAT) and a global region needing attention of partial Fourier algorithm; the artificial intelligence, namely neural network learning, only needs to predict and constrain local regions, and does not need to constrain outside the local regions. The parallel acquisition algorithm can be that firstly, the sensitivity information of the phased array coil of each point in the imaging tissue is obtained; then, low-density filling of K space phase encoding lines is carried out by utilizing fewer magnetic resonance signals acquired by the phased array coil; and finally, removing the curls by using the sensitivity information of the phased array coil and adopting a mathematical method to obtain images of all visual fields. Alternatively, the parallel acquisition algorithm may be, for example, a sensitivity encoding (SENSE) technique, a spatially tuned synchronous acquisition (SMASH) technique, or a self-calibrated automatic parallel acquisition (GRAPPA) technique. The partial fourier algorithm, for example, collects half or more than half of the phase encoding lines of the K space, and the remaining phase encoding lines that are not collected are depopulated according to the symmetry principle of the K space in the phase encoding direction.
In the magnetic resonance image reconstruction method provided by this embodiment, an initial image is obtained by preliminarily reconstructing the K-space data set, and iterative reconstruction is performed on the K-space data set and the initial image through different constraint algorithms based on the set constraints, so as to obtain a target image of the object. The reconstruction of the K space data set corresponds to the whole of the initial image, and the reconstruction of the initial image corresponds to the local of the initial image, so that different constraints can concern different areas when the constraint algorithm is fused and reconstructed, and the calculation efficiency is improved; the reconstruction of the initial image only corresponds to the local part of the initial image, and the overall reconstruction is not needed, so that the calculation efficiency is further improved, and the magnetic resonance imaging efficiency is improved. On the other hand, considering that different constraint algorithms have different sensitivities to image reconstruction, the constraint reconstruction with high sensitivity is selected for the interested region; and for the constrained reconstruction with low non-interested region selection sensitivity, the signal-to-noise ratio of the target image can be improved.
Referring to fig. 3, the present embodiment relates to a possible implementation manner of performing iterative reconstruction on the K-space data set and the initial image based on the set constraint to obtain the target image of the object, that is, S30 includes:
s310, establishing an image reconstruction model based on set constraints, wherein the set constraints comprise at least two types, a reconstruction region of one type of constraint comprises the global state of the initial image, and a reconstruction region of the other type of constraint only comprises the local state of the initial image;
and S320, performing iterative reconstruction on the K space data set and the initial image according to the image reconstruction model to obtain a target image.
Based on the set constraint, fusing multiple constraint algorithms, establishing an image reconstruction model, and performing iterative reconstruction based on the image reconstruction model to obtain a target image. The image reconstruction model is fused with at least two constraint algorithms, wherein one constraint algorithm corresponds to the whole part of the initial image, and the other constraint algorithm corresponds to the local part of the initial image. In this way, different constraint algorithms are adopted for different regions of the initial image of the object, so that each constraint algorithm only exerts respective constraint capability in the self constraint region, and the image reconstruction efficiency is further improved.
Referring to fig. 4, in one embodiment, the process of building an image reconstruction model based on the set constraints includes the following steps, i.e., S310 includes:
s311, setting weight parameters of each set constraint;
s312, establishing an image reconstruction model based on the weight parameters of the set constraints.
The selection and combination of the constraints in the image reconstruction model and the weight parameters of each constraint can be designed according to actual requirements. In one embodiment, the image reconstruction model may be:
min∑ρifi,j,k(x)
where x denotes the image pixel value of the target region, fi,j,k(x) Representing the i-th constraint, j representing the space in which the i-th constraint is performed, k representing the reconstruction region of the i-th constraint, piWeights representing said ith constraintAnd (4) heavy parameters.
The number of constraints is not limited, each constraint is different, the space for constraint execution can be different, the reconstruction region of the constraint can be different, and the weight parameters of the constraint can be different. The image reconstruction model provided by the embodiment fuses different constraint algorithms, and different constraints focus on different reconstruction regions according to own algorithms, so that each constraint can fully exert own constraint capability, and the final effect is restricted. The method provided by the embodiment improves the calculation efficiency and improves the quality of the reconstructed image.
In one embodiment, the set constraints include at least two of partial fourier constraints, parallel/parallel acquisition constraints, Compressed Sensing (CS) constraints, and artificial intelligence constraints.
In one embodiment, setting constraints includes at least merging constraints and artificial intelligence constraints. The reconstruction area restricted by the artificial intelligence can be global, namely the global of the initial image, and the reconstruction area restricted by the artificial intelligence can be only the interested area, namely the local of the initial image.
In one embodiment, the image reconstruction model may be:
min(PFK,T+PIH,T+CSI,T+AII,P)
wherein, PFK,TA partial Fourier constraint representing that a reconstruction region performed in K space is global of an initial image; PI (proportional integral)H,TRepresenting the reconstruction area in the K space and/or the image domain as the global and adopting constraint of the initial image; CSI,TA global compressed sensing constraint representing a reconstructed region in an image domain as an initial image; AII,PAnd representing artificial intelligence constraint that a reconstruction area performed in an image domain is local to the initial image, wherein the local of the initial image can correspond to the interested area.
Specifically, the image reconstruction model may be:
whereinX is data of the initial image; y is a K space data set obtained in an undersampling mode; the subscript T indicates that the initial image data all participate in the calculation, namely the initial image is subjected to global reconstruction; p represents that part of the initial image data participates in calculation, namely the initial image is locally reconstructed; f represents a Fourier transform constraint, namely performing Fourier transform operation on the image; g is K space parallel acquisition constraint, namely K space conjugate symmetry operation is carried out on the image; c is K space conjugate symmetry constraint, namely K space conjugate symmetry operation is carried out on the image; xe is data of initial image data x after phase calibration; CS is a compressed sensing constraint, namely, the compressed sensing operation is carried out on the image; AI is artificial intelligent constraint, namely neural network prediction is carried out on the image; lambda [ alpha ]iIs a weighting factor. In one embodiment, the weighting factor may be obtained by: lambda [ alpha ]iSi is the confidence factor for the ith constraint and Σ Sj is the overall confidence factor. When the above formula (1) satisfies the minimum value, x is the desired solution. In another embodiment, the weighting factor may also be determined by the signal-to-noise ratio of the reconstructed image under different constraints. For example, the image reconstruction model comprises a first image of a compressed sensing constraint globally corresponding to the initial image and a second image of an artificial intelligence constraint locally corresponding to the initial image; respectively calculating the SNR of the first image relative to the initial image1SNR of the second image relative to the initial image2(ii) a And determining a weight factor according to the signal-to-noise ratio obtained by calculation. For example, images with a high signal-to-noise ratio may be assigned a large weighting factor and images with a low signal-to-noise ratio may be assigned a small weighting factor. Illustratively, the weighting factor of the second image may be SNR2Ratio to the total signal-to-noise ratio of other reconstructed images. The following formula may be used:
wherein m represents the constraint type contained in the image reconstruction model, and m is more than or equal to 1; SNRiRepresenting the signal-to-noise ratio of the ith constrained reconstruction image relative to the initial image, wherein n is more than or equal to i and less than or equal to m; SNRnRepresenting what is contained in the reconstructed modelThe sum of the signal-to-noise ratios of the constrained reconstructed image relative to the original image.
In one embodiment, the image reconstruction model includes two constraint algorithms performed in the image domain, and the constraint algorithms respectively correspond to the local and global of the initial image, the locally reconstructed image corresponding to the initial image may provide a reference for the globally reconstructed image corresponding to the initial image, or the globally reconstructed image corresponding to the initial image may be based on the locally reconstructed image corresponding to the initial image. For example, an artificial intelligence constraint algorithm may be firstly adopted to reconstruct in an image domain to obtain a second image, where the second image is a local image; correcting the initial image by using the local image, for example, replacing a corresponding part of the initial image by using the local image; and reconstructing the corrected initial image domain by adopting a compressed sensing constraint algorithm to obtain a first image, so that the reconstructed local accuracy is improved.
In one embodiment, the target image is at least two, and the at least two kinds of reconstruction regions with set constraints respectively correspond to different parts of the initial image. For example, the target image includes a head image and a chest image. The reconstruction areas of the two constraints are respectively a head-related local area and a chest-related local area, each constraint only focuses on a part of the object, and the image information of different positions of the object is obtained by simultaneous calculation of multiple constraint algorithms, so that multiple target images are obtained, and the magnetic resonance imaging efficiency is further improved.
It should be understood that, although the steps in the flowchart are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a portion of the steps in the flowchart may include multiple sub-steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of performing the sub-steps or stages is not necessarily sequential, but may be performed alternately or alternately with other steps or at least a portion of the sub-steps or stages of other steps.
Referring to fig. 5, an embodiment of the present application further provides a magnetic resonance image reconstruction apparatus 10, which includes a data acquisition module 110, a preliminary reconstruction module 120, and an iterative reconstruction module 130, wherein:
a data acquisition module 110, configured to acquire a K-space data set corresponding to a magnetic resonance signal of a subject, where the K-space data set is acquired in an undersampling manner;
a preliminary reconstruction module 120, configured to perform preliminary reconstruction on the K-space data set to obtain an initial image;
an iterative reconstruction module 130, configured to perform iterative reconstruction on the K-space data set and the initial image based on a set constraint to obtain a target image of the object, where the reconstruction of the K-space data set corresponds to a global state of the initial image, and the reconstruction of the initial image corresponds to a local state of the initial image.
In one embodiment, the iterative reconstruction module 130 is specifically configured to build an image reconstruction model based on the set constraints, where the set constraints include at least two types, a reconstruction region of one type includes a global region of the initial image, and a reconstruction region of the other type includes only a local region of the initial image; and performing iterative reconstruction on the K space data set and the initial image according to the image reconstruction model to obtain the target image.
In one embodiment, the iterative reconstruction module 130 is further configured to set a weight parameter for each of the set constraints; and establishing the image reconstruction model based on the weight parameters of the set constraints.
In one embodiment, the image reconstruction model is: min Σ ρifi,j,k(x) Where x denotes the image pixel value of the target region, fi,j,k(x) Representing the i-th constraint, j representing the space in which the i-th constraint is performed, k representing the reconstruction region of the i-th constraint, piA weight parameter representing said ith constraint.
In one embodiment, the set constraints include at least two of a partial fourier constraint, a merge constraint, a compressive sensing constraint, and an artificial intelligence constraint.
In one embodiment, the number of the target images is at least two, and the at least two types of reconstruction regions with set constraints respectively correspond to different parts of the initial image.
In one embodiment, the image reconstruction model is:
For specific limitations of the magnetic resonance image reconstruction apparatus 10, reference may be made to the above description of the magnetic resonance image reconstruction method, which is not repeated herein. The respective modules in the magnetic resonance image reconstruction apparatus 10 described above may be wholly or partially implemented by software, hardware, and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
Referring to fig. 6, in one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 6. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing source data, report data and the like. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a magnetic resonance image reconstruction method.
Those skilled in the art will appreciate that the architecture shown in fig. 6 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:
acquiring a K space data set corresponding to a magnetic resonance signal of an object, wherein the K space data set is acquired in an undersampling mode;
performing primary reconstruction on the K space data set to obtain an initial image;
based on a set constraint, performing iterative reconstruction on the K-space data set and the initial image to obtain a target image of the object, wherein the reconstruction of the K-space data set corresponds to the whole of the initial image, and the reconstruction of the initial image corresponds to the local of the initial image.
In one embodiment, the processor, when executing the computer program, further performs the steps of: establishing an image reconstruction model based on the set constraints, wherein the set constraints comprise at least two types, the reconstruction region of one type of constraint comprises the global part of the initial image, and the reconstruction region of the other type of constraint comprises only the local part of the initial image; and performing iterative reconstruction on the K space data set and the initial image according to the image reconstruction model to obtain the target image.
In one embodiment, the processor, when executing the computer program, further performs the steps of: setting a weight parameter of each set constraint; and establishing the image reconstruction model based on the weight parameters of the set constraints.
In one embodiment, the image reconstruction model is: min Σ ρifi,j,k(x) Where x denotes the image pixel value of the target region, fi,j,k(x) Representing the i-th constraint, j representing the space in which the i-th constraint is performed, k representing the reconstruction region of the i-th constraint, piA weight parameter representing said ith constraint.
In one embodiment, the set constraints include at least two of a partial fourier constraint, a merge constraint, a compressive sensing constraint, and an artificial intelligence constraint.
In one embodiment, the number of the target images is at least two, and the at least two types of reconstruction regions with set constraints respectively correspond to different parts of the initial image.
In one embodiment, the image reconstruction model is:
In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:
acquiring a K space data set corresponding to a magnetic resonance signal of an object, wherein the K space data set is acquired in an undersampling mode;
performing primary reconstruction on the K space data set to obtain an initial image;
based on a set constraint, performing iterative reconstruction on the K-space data set and the initial image to obtain a target image of the object, wherein the reconstruction of the K-space data set corresponds to the whole of the initial image, and the reconstruction of the initial image corresponds to the local of the initial image.
In one embodiment, the computer program when executed by the processor further performs the steps of: establishing an image reconstruction model based on the set constraints, wherein the set constraints comprise at least two types, the reconstruction region of one type of constraint comprises the global part of the initial image, and the reconstruction region of the other type of constraint comprises only the local part of the initial image; and performing iterative reconstruction on the K space data set and the initial image according to the image reconstruction model to obtain the target image.
In one embodiment, the computer program when executed by the processor further performs the steps of: setting a weight parameter of each set constraint; and establishing the image reconstruction model based on the weight parameters of the set constraints.
In one embodiment, the image reconstruction model is: min Σ ρifi,j,k(x) Where x denotes the image pixel value of the target region, fi,j,k(x) Representing the i-th constraint, j representing the space in which the i-th constraint is performed, k representing the reconstruction region of the i-th constraint, piA weight parameter representing said ith constraint.
In one embodiment, the set constraints include at least two of a partial fourier constraint, a merge constraint, a compressive sensing constraint, and an artificial intelligence constraint.
In one embodiment, the number of the target images is at least two, and the at least two types of reconstruction regions with set constraints respectively correspond to different parts of the initial image.
In one embodiment, the image reconstruction model is:
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
The technical features of the embodiments described above may be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the embodiments described above are not described, but should be considered as being within the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the claims. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (10)
1. A magnetic resonance image reconstruction method, comprising:
acquiring a K space data set corresponding to a magnetic resonance signal of an object, wherein the K space data set is acquired in an undersampling mode;
performing primary reconstruction on the K space data set to obtain an initial image;
based on a set constraint, performing iterative reconstruction on the K-space data set and the initial image to obtain a target image of the object, wherein the reconstruction of the K-space data set corresponds to the whole of the initial image, and the reconstruction of the initial image corresponds to the local of the initial image.
2. The method of claim 1, wherein performing an iterative reconstruction of the K-space dataset and the initial image to obtain a target image of a subject based on a set constraint comprises:
establishing an image reconstruction model based on the set constraints, wherein the set constraints comprise at least two types, the reconstruction region of one type of constraint comprises the global part of the initial image, and the reconstruction region of the other type of constraint comprises only the local part of the initial image;
and performing iterative reconstruction on the K space data set and the initial image according to the image reconstruction model to obtain the target image.
3. The method of claim 2, wherein said building an image reconstruction model based on said set constraints comprises:
setting a weight parameter of each set constraint;
and establishing the image reconstruction model based on the weight parameters of the set constraints.
4. The method of claim 3, wherein the image reconstruction model is:
min∑ρifi,j,k(x)
where x represents the image pixel value of the target region, fi,j,k(x) Representing the i-th constraint, j representing the space in which the i-th constraint is performed, k representing the reconstruction region of the i-th constraint, piA weight parameter representing said ith constraint.
5. The method of claim 2, wherein the set constraints comprise at least two of a partial fourier constraint, a merge constraint, a compressive sensing constraint, and an artificial intelligence constraint.
6. The method according to claim 2, wherein the target image is at least two, and the at least two kinds of reconstruction regions with set constraints respectively correspond to different parts of the initial image.
7. The method according to claim 4 or 5, wherein the image reconstruction model is:
wherein x is data of the initial image, y is the K space data set, T is global reconstruction of the initial image, P is local reconstruction of the initial image, F is Fourier transformation constraint, G is K space parallel acquisition constraint, C is K space conjugate symmetry constraint, xe is data of the initial image after phase calibration, CS is compressive sensing constraint, AI is artificial intelligence constraint, and lambda is a linear transformation constraintiAs a weighting factor, λiSi is the confidence factor for the ith constraint.
8. A magnetic resonance image reconstruction apparatus, characterized by comprising:
the data acquisition module is used for acquiring a K space data set corresponding to a magnetic resonance signal of an object, wherein the K space data set is acquired in an undersampling mode;
the preliminary reconstruction module is used for carrying out preliminary reconstruction on the K space data set to obtain an initial image;
an iterative reconstruction module, configured to perform iterative reconstruction on the K-space data set and the initial image based on a set constraint to obtain a target image of the object, where the reconstruction of the K-space data set corresponds to a global state of the initial image, and the reconstruction of the initial image corresponds to a local state of the initial image.
9. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method of any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911235888.6A CN112927313B (en) | 2019-12-05 | 2019-12-05 | Magnetic resonance image reconstruction method, magnetic resonance image reconstruction device, computer equipment and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911235888.6A CN112927313B (en) | 2019-12-05 | 2019-12-05 | Magnetic resonance image reconstruction method, magnetic resonance image reconstruction device, computer equipment and readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112927313A true CN112927313A (en) | 2021-06-08 |
| CN112927313B CN112927313B (en) | 2022-12-06 |
Family
ID=76161646
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911235888.6A Active CN112927313B (en) | 2019-12-05 | 2019-12-05 | Magnetic resonance image reconstruction method, magnetic resonance image reconstruction device, computer equipment and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112927313B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113763499A (en) * | 2021-08-27 | 2021-12-07 | 上海壁仞智能科技有限公司 | Nuclear magnetic resonance image reconstruction method, apparatus, device, medium, and program product |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| US20140091793A1 (en) * | 2012-09-28 | 2014-04-03 | Tsinghua University | Method for diffusion magnetic resonance imaging |
| CN104434108A (en) * | 2014-12-09 | 2015-03-25 | 中国科学院武汉物理与数学研究所 | Multi-layer magnetic resonance fast imaging method based on adjacent layer information and undersampling |
| CN104899906A (en) * | 2015-06-12 | 2015-09-09 | 南方医科大学 | Magnetic resonance image reconstruction method based on adaptive orthogonal basis |
| CN109325985A (en) * | 2018-09-18 | 2019-02-12 | 上海联影智能医疗科技有限公司 | MR image reconstruction method, apparatus and computer readable storage medium |
| CN110148215A (en) * | 2019-05-22 | 2019-08-20 | 哈尔滨工业大学 | A kind of four-dimensional MR image reconstruction method based on smoothness constraint and local low-rank restricted model |
-
2019
- 2019-12-05 CN CN201911235888.6A patent/CN112927313B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140091793A1 (en) * | 2012-09-28 | 2014-04-03 | Tsinghua University | Method for diffusion magnetic resonance imaging |
| 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 |
| CN104434108A (en) * | 2014-12-09 | 2015-03-25 | 中国科学院武汉物理与数学研究所 | Multi-layer magnetic resonance fast imaging method based on adjacent layer information and undersampling |
| CN104899906A (en) * | 2015-06-12 | 2015-09-09 | 南方医科大学 | Magnetic resonance image reconstruction method based on adaptive orthogonal basis |
| CN109325985A (en) * | 2018-09-18 | 2019-02-12 | 上海联影智能医疗科技有限公司 | MR image reconstruction method, apparatus and computer readable storage medium |
| CN110148215A (en) * | 2019-05-22 | 2019-08-20 | 哈尔滨工业大学 | A kind of four-dimensional MR image reconstruction method based on smoothness constraint and local low-rank restricted model |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113763499A (en) * | 2021-08-27 | 2021-12-07 | 上海壁仞智能科技有限公司 | Nuclear magnetic resonance image reconstruction method, apparatus, device, medium, and program product |
| CN113763499B (en) * | 2021-08-27 | 2023-08-18 | 上海壁仞智能科技有限公司 | Nuclear magnetic resonance image reconstruction method, device, equipment and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112927313B (en) | 2022-12-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chung et al. | Score-based diffusion models for accelerated MRI | |
| Wang et al. | DeepcomplexMRI: Exploiting deep residual network for fast parallel MR imaging with complex convolution | |
| Liu et al. | MANTIS: model‐augmented neural network with incoherent k‐space sampling for efficient MR parameter mapping | |
| Hammernik et al. | Systematic evaluation of iterative deep neural networks for fast parallel MRI reconstruction with sensitivity‐weighted coil combination | |
| US10996306B2 (en) | MRI system and method using neural network for detection of patient motion | |
| CN114450599B (en) | Maxwell Wei Binghang imaging | |
| KR102584166B1 (en) | MAGNETIC RESONANCE IMAGE PROCESSING APPARATUS AND METHOD USING ARTIFICIAL NEURAL NETWORK AND RESCAlING | |
| Oh et al. | ETER-net: end to end MR image reconstruction using recurrent neural network | |
| JP2020146449A (en) | Magnetic resonance image high-speed reconfiguring method and magnetic resonance imaging device | |
| CN111257810B (en) | Magnetic resonance signal acquisition system and method | |
| CN111157935B (en) | Magnetic resonance imaging method, magnetic resonance imaging device, storage medium and computer equipment | |
| US11948676B2 (en) | Qualitative and quantitative MRI using deep learning | |
| CN115639510A (en) | Magnetic resonance imaging method, spectroscopic imaging method, apparatus, device, and storage medium | |
| CN112927313B (en) | Magnetic resonance image reconstruction method, magnetic resonance image reconstruction device, computer equipment and readable storage medium | |
| CN115222628A (en) | Image processing method, device, equipment and storage medium | |
| Hu et al. | SPICE: Self-supervised learning for MRI with automatic coil sensitivity estimation | |
| CN113050009B (en) | Three-dimensional magnetic resonance rapid parameter imaging method and device | |
| CN114494014A (en) | Magnetic resonance image super-resolution reconstruction method and device | |
| Angshul | A sparsity based approach towards fast MRI acquisition | |
| US11965947B2 (en) | Multiplex MRI image reconstruction | |
| CN112578325B (en) | Magnetic resonance imaging method, magnetic resonance imaging apparatus, computer device, and storage medium | |
| CN114062988B (en) | Magnetic resonance spectrum imaging method, apparatus, computer device and storage medium | |
| WO2022124473A1 (en) | Magnetic resonance image processing apparatus with artificial neural network applied to k-space domain and method therefor | |
| Liu et al. | MANTIS: Model-Augmented Neural neTwork with Incoherent k-space Sampling for efficient MR T2 mapping | |
| US20230014745A1 (en) | Deep learning based image reconstruction |
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 |