CN110338795B - Radial golden angle magnetic resonance heart film imaging method, device and equipment - Google Patents
Radial golden angle magnetic resonance heart film imaging method, device and equipment Download PDFInfo
- Publication number
- CN110338795B CN110338795B CN201910625835.9A CN201910625835A CN110338795B CN 110338795 B CN110338795 B CN 110338795B CN 201910625835 A CN201910625835 A CN 201910625835A CN 110338795 B CN110338795 B CN 110338795B
- Authority
- CN
- China
- Prior art keywords
- network
- space data
- image
- undersampled
- frequency domain
- 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
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/05—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves
- A61B5/055—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves involving electronic [EMR] or nuclear [NMR] magnetic resonance, e.g. magnetic resonance imaging
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7225—Details of analog processing, e.g. isolation amplifier, gain or sensitivity adjustment, filtering, baseline or drift compensation
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7253—Details of waveform analysis characterised by using transforms
- A61B5/7257—Details of waveform analysis characterised by using transforms using Fourier transforms
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
-
- 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 application discloses a radial golden angle magnetic resonance heart movie imaging method, a radial golden angle magnetic resonance heart movie imaging device and radial golden angle magnetic resonance heart movie imaging equipment. The method comprises the following steps: acquiring magnetic resonance K space data of radial golden angle sampling of the preset channel number of a target object; rearranging the K space data to obtain undersampled K space data with preset frames; and simultaneously inputting the undersampled K space data of the preset frame number into a cross domain image reconstruction network to acquire a target image. The embodiment of the application solves the problem of long time consumption in the image reconstruction process based on radial golden angle sampling data; the method can shorten the time for reconstructing the image of the heart magnetic resonance data based on radial golden angle sampling and improve the quality of the reconstructed image.
Description
Technical Field
The embodiment of the application relates to a medical imaging technology, in particular to a radial golden angle magnetic resonance heart movie imaging method, a radial golden angle magnetic resonance heart movie imaging device and radial golden angle magnetic resonance heart movie imaging equipment.
Background
Magnetic resonance cardiac cine imaging is a non-invasive imaging technique that can be used to assess cardiac function, wall motion abnormalities, etc., providing rich information for cardiac clinical diagnosis. However, due to the physical and hardware limitations of magnetic resonance and the duration of cardiac motion cycles, magnetic resonance cardiac cine imaging tends to be limited in time and spatial resolution, and it is not possible to accurately assess cardiac function conditions of a portion of cardiac diseases, such as arrhythmia. Therefore, on the premise of ensuring imaging quality, it is important to improve the time and spatial resolution of magnetic resonance cardiac cine imaging by using a rapid imaging method.
Currently, commonly used methods for accelerating magnetic resonance cardiac cine Imaging, including Parallel Imaging (PI) and compressed sensing (Compressed Sensing, CS) techniques, utilize spatial information of the data to fill in undersampled K-space data. To obtain higher acceleration multiples, sparse sensing techniques combining compressed sensing and parallel imaging are proposed. The radial golden angle sparse parallel technology expands the idea to a radial golden angle sampling mode and is successfully applied to the fields of free breathing abdomen magnetic resonance imaging, pediatric body magnetic resonance imaging, breast magnetic resonance imaging, neck magnetic resonance imaging and the like. However, radial golden angle sparse parallelism still suffers from some degree of motion blur, especially in patients or elderly. The motion state is reconstructed by the motion state dimension-radial golden angle sparse parallel technology, so that motion blur is effectively relieved.
However, the motion state dimension-radial golden angle sparse parallel technology processes rearranged data frames frame by frame, the imaging process is slower, and the time is long.
Disclosure of Invention
The embodiment of the application provides a radial golden angle magnetic resonance heart movie imaging method, device and equipment, which are used for shortening the time for reconstructing images of heart magnetic resonance data based on radial golden angle sampling and improving the quality of reconstructed images.
In a first aspect, an embodiment of the present application provides a radial golden angle magnetic resonance cardiac cine imaging method, the method including:
acquiring magnetic resonance K space data of radial golden angle sampling of the preset channel number of a target object;
rearranging the K space data to obtain undersampled K space data with preset frames;
and simultaneously inputting the undersampled K space data of the preset frame number into a cross domain image reconstruction network to acquire a target image.
In a second aspect, an embodiment of the present application further provides a radial golden angle magnetic resonance cardiac cine imaging apparatus, the apparatus comprising:
the data acquisition module is used for acquiring magnetic resonance K space data of radial golden angle sampling of the preset channel number of the target object;
the data preprocessing module is used for rearranging the K space data to obtain undersampled K space data with preset frames;
and the image reconstruction module is used for simultaneously inputting the undersampled K space data of the preset frame number into a cross domain image reconstruction network so as to acquire a target image.
In a third aspect, embodiments of the present application further provide a magnetic resonance scanning system, the magnetic resonance scanning system comprising: scanning equipment, treatment couch and computer equipment;
wherein the computer device comprises:
the system comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor is used for realizing the radial golden angle magnetic resonance cardiac cine imaging method according to any embodiment of the application when the processor executes the computer program.
In a fourth aspect, embodiments of the present application further provide a computer readable storage medium having stored thereon a computer program which when executed by a processor implements a radial golden angle magnetic resonance cardiac cine imaging method according to any one of the embodiments of the present application.
According to the embodiment of the application, the acquired magnetic resonance K space data sampled at the radial golden angle of the preset channel number of the target object is rearranged to obtain undersampled K space data with the preset frame number, and then the undersampled K space data is input into a cross domain image reconstruction network to obtain a reconstructed target image, so that the problem that the time for reconstructing the image of the magnetic resonance data sampled at the radial golden angle is long in the prior art is solved; the method can shorten the time for reconstructing the image of the heart magnetic resonance data based on radial golden angle sampling and improve the quality of the reconstructed image.
Drawings
FIG. 1 is a flow chart of a radial golden angle magnetic resonance cardiac cine imaging method in accordance with a first embodiment of the present application;
FIG. 2 is a schematic view of a radial golden angle acquisition trajectory in accordance with a first embodiment of the present application;
FIG. 3 is a schematic diagram of a cross-domain image reconstruction network according to a first embodiment of the present application;
fig. 4 is a diagram showing an example of the structure of a first frequency domain sub-network according to the first embodiment of the present application;
fig. 5 is a diagram showing an example of the structure of a second frequency domain sub-network according to the first embodiment of the present application;
FIG. 6 is a schematic diagram of a radial golden angle MRI apparatus in accordance with a second embodiment of the present application;
fig. 7 is a schematic structural diagram of a magnetic resonance system in a third embodiment of the present application;
fig. 8 is a schematic structural diagram of a computer device in a magnetic resonance system according to a third embodiment of the present application.
Detailed Description
The application is described in further detail below with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the application and are not limiting thereof. It should be further noted that, for convenience of description, only some, but not all of the structures related to the present application are shown in the drawings.
Example 1
Fig. 1 is a flowchart of a radial golden angle magnetic resonance cardiac cine imaging method according to an embodiment of the present application, where the method may be implemented by a radial golden angle magnetic resonance cardiac cine imaging apparatus according to radial golden angle sampling data, and may specifically be implemented by software and/or hardware in an electronic device, where the electronic device may be a magnetic resonance scanner.
As shown in fig. 1, the radial golden angle magnetic resonance cardiac cine imaging method specifically includes:
s110, acquiring magnetic resonance K space data of radial golden angle sampling of the preset channel number of the target object.
The target object may be a human or animal, and is a scanning target object of the magnetic resonance scanner when diagnosis of heart diseases or evaluation of heart functions is required.
When a magnetic resonance scanner acquires data, a plurality of coils usually acquire data at the same time, so as to acquire heart magnetic resonance data of different angles of a target object, and the number of the coils is determined according to the setting of the magnetic resonance scanner. Specifically, the coils correspond to a plurality of channels, namely, the magnetic resonance scanner collects multichannel K space data, the K space is also called Fourier space, and is a filling space of original digital data of magnetic resonance signals with space positioning coding information, and each magnetic resonance image has a corresponding K space data lattice. When the acquired data are the sampling data of the radial golden angle, the magnetic resonance K space data of the radial golden angle sampling of the preset channel number of the target object can be obtained.
In radial acquisition, the sampling trajectory is typically taken from the center of k-space to the k-space edge or from one edge of k-space to the other, and in the present application, the second sampling trajectory is exemplified. The radial golden angle data acquisition track can refer to the radial golden angle acquisition track schematic diagram shown in fig. 2, and the radial golden angle acquisition is rotated by a fixed angle of 111.25 degrees each time, and the value is calculated based on the golden section ratio, and is called as a golden angle. The radial golden angle acquisition of data has the advantage that k-space at any time during the sampling process is rendered in a relatively uniform state, which is an image reconstruction for data using any number of sampling lines.
S120, rearranging the K space data to obtain undersampled K space data with preset frames.
Specifically, the K space data is rearranged to obtain undersampled K space data with a preset frame number, which may be that the K space data is averagely distributed, so that each frame of undersampled K space data contains the K space data corresponding to the same number of sampling lines. For example, assuming that there are 256 sample data on each sample line, the sample data of each sample line are arranged in a column manner, and there are 1000 sample lines in total, then all the sample data after arrangement have 256×1000 data, and if the data of each 20 sample lines are divided to obtain undersampled K-space data, then 50 frames of undersampled K-space data of 256×20 can be obtained.
In another preferred embodiment, the K-space data is rearranged to obtain undersampled K-space data with a preset frame number, and the K-space data may be unevenly divided into undersampled K-space data with a preset frame number according to a motion state of the target object, where the motion state includes a respiratory frequency or a heartbeat frequency of the target object. For example, in the process of acquiring data, the heart of the target object contracts and expands along with the rhythm of respiration, if 30 pieces of sampling line data are acquired during systole and 20 pieces of sampling line data are acquired during diastole, 256×30 pieces of data can be arranged into one frame of undersampled K-space data, 256×20 pieces of data are arranged into one frame of undersampled K-space data, and then the data arrangement process is repeated until all undersampled K-space data are acquired. And the data rearrangement is carried out on the motion state according to the target object, so that noise such as image artifacts and the like caused by motion blur is effectively relieved.
S130, undersampling K space data of the preset frame number is simultaneously input into a cross domain image reconstruction network to obtain a target image.
Specifically, the cross-domain image reconstruction network includes: a frequency domain sub-network and an image domain sub-network, wherein the frequency domain sub-network comprises a first frequency domain sub-network and a second frequency domain sub-network. The specific structure of the cross-domain image reconstruction network can be seen with reference to fig. 3.
The image reconstruction process through the cross domain image reconstruction network mainly comprises the following steps: firstly, the undersampled K space data of the preset frame number is input into a first frequency domain sub-network according to a first preset format to obtain first characteristic information of the undersampled K space data of the preset frame number on a channel-space, and meanwhile, the undersampled K space data of the preset frame number is input into a second frequency domain sub-network according to a second preset format to obtain second characteristic information of the undersampled K space data of the preset frame number on a time-space. Wherein K1 is an undersampled K-space format of a first preset format, and the first preset format is (kt, kx, ky, coil), so as to enable the first frequency domain sub-network to convolve in the (kx, ky, coil) dimension, so that the first frequency domain sub-network extracts channel-space information. K2 is an undersampled K-space format of a second preset format, and the second preset format is (coil, kx, ky, kt), so as to enable the second frequency domain sub-network to convolve in the (kx, ky, kt) dimension, so that the second frequency domain sub-network extracts time-space information. Therefore, the two sub-networks can independently learn the characteristics in different dimensions, which is beneficial to fully exploring the correlation of parallel heart dynamic data and mining the redundancy of the data in space, time and channels as much as possible.
Specifically, the first frequency domain sub-network may use a residual density network (Residual Dense Network, RDN), which may be structured as shown in fig. 4. The network has the greatest characteristic that local and global features with different depths can be fused, so that the features of the whole network are effectively utilized. The residual density network mainly consists of five major parts, namely: 1) Shallow layer feature extraction; 2) A residual density module (Residual Dense Block, RDB) for local feature fusion; 3) Global feature fusion; 4) Global residual error learning; 5) Deep layer feature extraction.
The forward procedure of the first frequency domain sub-network is as follows:
step 0— input: under-sampling multi-channel K space data, wherein the dimension of the K space data is (kt, kx, ky, coil). Setting the K-space data dimension to (kt, kx, ky, coil) may cause the 3D convolution layer to convolve on (kx, ky, coil), i.e., cause the present sub-network to learn channel-space characteristics.
Step 1, shallow feature extraction: is composed of two 3D convolution layers. The undersampled multi-channel K space data firstly passes through the two convolution layers to extract shallow layer characteristics.
Step 2, local feature fusion: consists of D residual density modules. There are several 3D convolution layers in each residual density module. Unlike common convolution layers, the output of each convolution layer is input not only to the next convolution layer, but also to all subsequent convolution layers of the RDB, constituting a density connection. The purpose of the density connection is: so that both deep and shallow convolution layers function. The outputs of these 3D convolutional layers are connected (con-cate) together and passed through a 1 x 1 convolutional layer for feature fusion. Thus, feature fusion of a density network is accomplished. And then the output of the residual density module is obtained through residual connection. Since each residual density module belongs to a part of the whole frequency domain sub-network, the feature fusion here is also called local feature fusion. Wherein D is an integer greater than zero
Step 3- -Global feature fusion: the outputs of the D residual density modules are connected (con-cate) together, and then a convolution layer of 1 x 1 is passed, so that the result of global feature fusion can be obtained.
Step 4, global residual learning: and (3) connecting the result obtained after the global feature fusion in the step (3) with a residual error through a 3D convolution layer, and performing global residual error learning.
Step 5, deep feature extraction: is composed of two 3D convolution layers. And (3) carrying out deep feature extraction on the result obtained after the global residual error learning through 2 3D convolution layers.
Step 6-outputting the multi-channel K space data (namely the extracted channel-space information and the first characteristic information) reconstructed by the first frequency domain sub-network.
And performing inverse Fourier transform (IFFT) on the reconstructed multi-channel K space data to obtain a multi-channel image. After the multichannel image is subjected to channel fusion (coil combination), a single-channel image (i.e., a first reconstructed image) can be obtained and used for extracting the characteristics of the image domain network. The specific method of channel fusion can be as follows: the multi-channel image points are multiplied by the conjugate of their channel sensitivities and summed in the channel direction.
Fig. 5 is a diagram showing an example of the structure of a second frequency domain sub-network, which is formed by cascading M convolution modules (i.e., frequency domain modules). In particular, the second frequency domain sub-network comprises M frequency domain modules (Fnet m M=1, …, M), each frequency domain module contains L3-dimensional convolutional layers (3D Conv) and one frequency domain data consensus layer (Kspace Data Consistency, KDC). The input to the second frequency domain sub-network is undersampled multi-channel K-space data, and setting the K-space dimension to (coil, kx, ky, kt) may cause the 3D convolution layer to convolve on (kx, ky, kt), i.e., cause the present sub-network to learn time-space characteristics. The forward procedure of the second frequency domain subnetwork may be expressed by the following formula:
first frequency domain module (m=1):
subsequent frequency domain modules (m=2, …, M):
the KDC is used for executing a frequency domain data consistency operation:
the convolution kernel and the offset of the first convolution layer in the mth frequency domain module are respectively; l=1, …, L; m=1, …, M; />Is the output of the ith convolutional layer in the mth frequency domain block. All but the last convolution layer is activated by a nonlinear function delta. After the characteristic is extracted by the convolution layer, correcting K space predicted by the network by utilizing the frequency domain data consistency layer, < >>Is to->And (5) correcting the result. Let the set of all acquired K-space coordinates be Ω. If the K-space coordinates (K x ,k y ) Within the set Ω ∈ ->Click-through of K-space through true acquisitionAnd (5) correcting the rows. Lambda is used to control the degree of data consistency, if lambda → infinity, the actual sampling point can be directly replacedCorresponding points. The embodiment of the application takes lambda-infinity. K (K) u A K-space data processed for the frequency domain network.
The final output of the second frequency domain sub-network isFor->The inverse fourier transform and channel fusion are performed to obtain single channel image data (i.e., a second reconstructed image).
Then, fusing the first reconstructed image and the second reconstructed image, and inputting the fused image into the image domain sub-network; and extracting image domain features of the fused image through the image domain sub-network, and obtaining a target image.
Optionally, the image domain network includes N image domain modules, each image domain module includes L three-dimensional convolution layers, a residual connection, and an image domain data consistency layer, where N is an integer greater than zero and L is an integer greater than zero.
In an embodiment of the present application, the image domain network may refer to a DC-CNN structure. The forward process of the image domain network can be expressed by the following formula:
subsequent image domain modules (n=2, …, N):
wherein the image domain data coherence (Image Data Consistency, IDC) is used to perform an image domain data coherence operation:
the convolution kernel and the offset term of the first convolution layer in the nth image domain module, i=1, …, L; n=1, …, N; />Is the output of the ith convolution layer in the nth image domain module. All but the last convolution layer is activated by a nonlinear function delta. After extracting features through a convolution layer, introducing residual error learning, S n Is the result of residual learning. For S n An image domain data consensus operation (IDC) is performed. IDC more switches between the frequency domain and the image domain than KDC. />Is to S n And performing an IDC image.
According to the technical scheme, the acquired magnetic resonance K space data of the radial golden angle sampling of the preset channel number of the target object are rearranged to obtain undersampled K space data of the preset frame number, then the undersampled K space data are input into a cross domain image reconstruction network to obtain a reconstructed target image, and the problem that image reconstruction of the magnetic resonance data of the radial golden angle sampling in the prior art is long in time consumption is solved; the method can shorten the time for reconstructing the image of the heart magnetic resonance data based on radial golden angle sampling and improve the quality of the reconstructed image.
Example two
Fig. 6 is a schematic structural diagram of a radial golden angle magnetic resonance cardiac cine imaging apparatus according to a second embodiment of the present application, where the apparatus may be configured in a magnetic resonance scanner in the case of performing magnetic resonance cardiac cine imaging according to radial golden angle sampling data.
As shown in fig. 6, a radial golden angle magnetic resonance cardiac cine imaging apparatus provided in an embodiment of the present application includes: a data acquisition module 610, a data preprocessing module 620, and an image reconstruction module 630.
The data acquisition module 610 is configured to acquire magnetic resonance K-space data of radial golden angle sampling of a preset number of channels of a target object; the data preprocessing module 620 is configured to rearrange the K-space data to obtain undersampled K-space data with a preset frame number; the image reconstruction module 630 is configured to input the undersampled K-space data of the preset frame number to a cross-domain image reconstruction network at the same time, so as to obtain a target image.
According to the technical scheme, the acquired magnetic resonance K space data of the radial golden angle sampling of the preset channel number of the target object are rearranged to obtain undersampled K space data of the preset frame number, then the undersampled K space data are input into a cross domain image reconstruction network to obtain a reconstructed target image, and the problem that image reconstruction of the magnetic resonance data of the radial golden angle sampling in the prior art is long in time consumption is solved; the method can shorten the time for reconstructing the image of the heart magnetic resonance data based on radial golden angle sampling and improve the quality of the reconstructed image.
Optionally, the data preprocessing module 620 is specifically configured to:
and carrying out average distribution on the K space data, so that each frame of undersampled K space data comprises the K space data corresponding to the same number of sampling lines.
Optionally, the data preprocessing module 620 may further be configured to:
and the K space data are unevenly divided into undersampled K space data with preset frames according to the motion state of the target object, wherein the motion state comprises the respiratory frequency or the heartbeat frequency of the target object.
Further, the cross-domain image reconstruction network includes: a frequency domain sub-network and an image domain sub-network, wherein the frequency domain sub-network comprises a first frequency domain sub-network and a second frequency domain sub-network.
Accordingly, the image reconstruction module 630 is specifically configured to:
inputting the undersampled K space data of the preset frame number into a first frequency domain sub-network according to a first preset format to obtain first characteristic information of the undersampled K space data of the preset frame number on a channel-space, and simultaneously inputting the undersampled K space data of the preset frame number into a second frequency domain sub-network according to a second preset format to obtain second characteristic information of the undersampled K space data of the preset frame number on a time-space;
respectively carrying out image reconstruction according to the first characteristic information and the second characteristic information to obtain a first reconstructed image and a second reconstructed image;
fusing the first reconstructed image and the second reconstructed image, and inputting the fused image into the image domain sub-network;
and extracting image domain features of the fused image through the image domain sub-network, and obtaining a target image.
Optionally, the first frequency domain sub-network is a residual density network, and the second frequency domain sub-network includes M frequency domain modules, each frequency domain module includes L three-dimensional convolution layers and a frequency domain data consistency layer, where M is an integer greater than zero, and L is an integer greater than zero.
Optionally, the image domain network includes N image domain modules, each image domain module includes L three-dimensional convolution layers, a residual connection, and an image domain data consistency layer, where N is an integer greater than zero and L is an integer greater than zero.
The radial golden angle magnetic resonance heart film imaging device provided by the embodiment of the application can execute the radial golden angle magnetic resonance heart film imaging method provided by any embodiment of the application, and has the corresponding functional modules and beneficial effects of the execution method.
Example III
Fig. 7 is a schematic structural diagram of a magnetic resonance scanning system in a third embodiment of the present application, the magnetic resonance scanning system including: scanning equipment, treatment couch and computer equipment.
The scanning equipment is used for acquiring magnetic resonance scanning data; a treatment couch user carries a target subject that receives the magnetic resonance scan and moves the target subject to the designated scan position; the computer equipment is used for controlling the working processes of the scanning equipment and the treatment bed so as to complete magnetic resonance scanning, and can also be used for acquiring magnetic resonance scanning data and processing the data so as to obtain a reconstructed target image.
Further, reference is made to fig. 8 for a schematic hardware structure of the computer device.
Fig. 8 is a schematic structural diagram of a computer device in a third embodiment of the present application. FIG. 8 illustrates a block diagram of an exemplary computer device 812 suitable for use in implementing embodiments of the present application. The computer device 812 shown in fig. 8 is merely an example and should not be construed as limiting the functionality and scope of use of embodiments of the present application. As shown in FIG. 8, computer device 812 is in the form of a general purpose computing device. Components of computer device 812 may include, but are not limited to: one or more processors or processing units 816, a system memory 828, and a bus 818 that connects the various system components, including the system memory 828 and the processing unit 816.
Bus 818 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus architectures. By way of example, and not limitation, such architectures include Industry Standard Architecture (ISA) bus, micro channel architecture (MAC) bus, enhanced ISA bus, video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus.
Computer device 812 typically includes a variety of computer system readable media. Such media can be any available media that is accessible by computer device 812 and includes both volatile and nonvolatile media, removable and non-removable media.
The system memory 828 may include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM) 830 and/or cache memory 832. Computer device 812 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage system 834 may be used to read from or write to non-removable, nonvolatile magnetic media (not shown in FIG. 8, commonly referred to as a "hard disk drive"). Although not shown in fig. 8, a magnetic disk drive for reading from and writing to a removable non-volatile magnetic disk (e.g., a "floppy disk"), and an optical disk drive for reading from or writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In such cases, each drive may be coupled to bus 818 through one or more data medium interfaces. Memory 828 may include at least one program product having a set (e.g., at least one) of program modules configured to carry out the functions of embodiments of the application.
A program/utility 840 having a set (at least one) of program modules 842 may be stored in, for example, memory 828, such program modules 842 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each or some combination of which may include an implementation of a network environment. Program modules 842 generally perform the functions and/or methods in the embodiments described herein.
The computer device 812 may also be in communication with one or more external devices 814 (e.g., keyboard, pointing device, display 824, etc.), wherein the display 824 may be used to display information corresponding to the digital radiography system and provide access to interact with the digital radiography system. The computer device 812 may also communicate with one or more devices that enable a user to interact with the computer device 812, and/or with any devices (e.g., network cards, modems, etc.) that enable the computer device 812 to communicate with one or more other computing devices. Such communication may occur through an input/output (I/O) interface 822. Moreover, computer device 812 may also communicate with one or more networks such as a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network, such as the Internet, through network adapter 820. As shown, the network adapter 820 communicates with other modules of the computer device 812 over the bus 818. It should be appreciated that although not shown in fig. 8, other hardware and/or software modules may be used in connection with computer device 812, including, but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, data backup storage systems, and the like.
The processing unit 816 executes various functional applications and data processing by running a program stored in the system memory 828, for example, to implement a radial golden angle magnetic resonance cardiac cine imaging method provided by an embodiment of the present application, the method mainly comprising:
acquiring magnetic resonance K space data of radial golden angle sampling of the preset channel number of a target object;
rearranging the K space data to obtain undersampled K space data with preset frames;
and simultaneously inputting the undersampled K space data of the preset frame number into a cross domain image reconstruction network to acquire a target image.
Example IV
The fourth embodiment of the present application also provides a computer readable storage medium having a computer program stored thereon, the program when executed by a processor implementing a radial golden angle magnetic resonance cardiac cine imaging method as provided by the fourth embodiment of the present application, the method mainly comprising:
acquiring magnetic resonance K space data of radial golden angle sampling of the preset channel number of a target object;
rearranging the K space data to obtain undersampled K space data with preset frames;
and simultaneously inputting the undersampled K space data of the preset frame number into a cross domain image reconstruction network to acquire a target image.
The computer storage media of embodiments of the application may take the form of any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
The computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, either in baseband or as part of a carrier wave. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Computer program code for carrying out operations of the present application may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, smalltalk, C ++ and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider).
Note that the above is only a preferred embodiment of the present application and the technical principle applied. It will be understood by those skilled in the art that the present application 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 application. Therefore, while the application has been described in connection with the above embodiments, the application is not limited to the embodiments, but may be embodied in many other equivalent forms without departing from the spirit or scope of the application, which is set forth in the following claims.
Claims (8)
1. A radial golden angle magnetic resonance cardiac cine imaging method, comprising:
acquiring magnetic resonance K space data of radial golden angle sampling of the preset channel number of a target object;
rearranging the K space data to obtain undersampled K space data with preset frames;
simultaneously inputting the undersampled K space data of the preset frame number into a cross domain image reconstruction network to obtain a target image;
the cross-domain image reconstruction network includes: a frequency domain sub-network and an image domain sub-network, wherein the frequency domain sub-network comprises a first frequency domain sub-network and a second frequency domain sub-networkThe method comprises the steps of carrying out a first treatment on the surface of the The first frequency domain sub-network is a residual density network, and the second frequency domain sub-network comprisesEach frequency domain module comprises +.>A three-dimensional convolution layer and a frequency domain data coincidence layer, wherein->Is an integer greater than zero, ">Is an integer greater than zero;
the step of simultaneously inputting the undersampled K-space data of the preset frame number into a cross domain image reconstruction network comprises the following steps:
inputting the undersampled K space data of the preset frame number into a first frequency domain sub-network according to a first preset format to obtain first characteristic information of the undersampled K space data of the preset frame number on a channel-space, and simultaneously inputting the undersampled K space data of the preset frame number into a second frequency domain sub-network according to a second preset format to obtain second characteristic information of the undersampled K space data of the preset frame number on a time-space, wherein the first preset format is (kt, kx, ky, coil), and the second preset format is (coil, kx, ky, kt).
2. The method of claim 1, wherein rearranging the K-space data to obtain undersampled K-space data of a preset frame number, comprises:
and carrying out average distribution on the K space data, so that each frame of undersampled K space data comprises the K space data corresponding to the same number of sampling lines.
3. The method of claim 1, wherein rearranging the K-space data to obtain undersampled K-space data of a preset frame number, comprises:
and the K space data are unevenly divided into undersampled K space data with preset frames according to the motion state of the target object, wherein the motion state comprises the respiratory frequency or the heartbeat frequency of the target object.
4. The method of claim 1, wherein simultaneously inputting the preset number of frames of undersampled K-space data to a cross-domain image reconstruction network to obtain a target image comprises:
respectively carrying out image reconstruction according to the first characteristic information and the second characteristic information to obtain a first reconstructed image and a second reconstructed image;
fusing the first reconstructed image and the second reconstructed image, and inputting the fused image into the image domain sub-network;
and extracting image domain features of the fused image through the image domain sub-network, and obtaining a target image.
5. The method of claim 1 or 4, wherein the image domain subnetwork comprisesEach image domain module comprising +.>A three-dimensional convolution layer, a residual connection, and an image domain data consensus layer, wherein,is an integer greater than zero, ">Is an integer greater than zero.
6. A radial golden angle magnetic resonance cardiac cine imaging apparatus, comprising:
the data acquisition module is used for acquiring magnetic resonance K space data of radial golden angle sampling of the preset channel number of the target object;
the data preprocessing module is used for rearranging the K space data to obtain undersampled K space data with preset frames;
the image reconstruction module is used for simultaneously inputting the undersampled K space data of the preset frame number into a cross domain image reconstruction network so as to acquire a target image; the cross-domain image reconstruction network includes: a frequency domain sub-network and an image domain sub-network, wherein the frequency domain sub-network comprises a first frequency domain sub-network and a second frequency domain sub-network; the first frequency domain sub-network is a residual density network, and the second frequency domain sub-network comprisesEach frequency domain module comprises +.>A three-dimensional convolution layer and a frequency domain data coincidence layer, wherein->Is an integer greater than zero, ">Is an integer greater than zero; the step of simultaneously inputting the undersampled K-space data of the preset frame number into a cross domain image reconstruction network comprises the following steps: inputting the undersampled K space data of the preset frame number into a first frequency domain sub-network according to a first preset format to obtain first characteristic information of the undersampled K space data of the preset frame number on a channel-space, and simultaneously inputting the undersampled K space data of the preset frame number into a second frequency domain sub-network according to a second preset format to obtain second characteristic information of the undersampled K space data of the preset frame number on a time-space, wherein the first preset format is (kt, kx, ky, coil), and the second preset format is (coil, kx, ky, kt).
7. A magnetic resonance scanning system, the magnetic resonance scanning system comprising:
scanning equipment, treatment couch and computer equipment;
wherein the computer device comprises: memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the radial golden angle magnetic resonance cardiac cine imaging method according to any one of claims 1 to 5 when the computer program is executed.
8. A computer readable storage medium storing a computer program, characterized in that the computer program when executed by a processor implements the radial golden angle magnetic resonance cardiac cine imaging method of any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910625835.9A CN110338795B (en) | 2019-07-11 | 2019-07-11 | Radial golden angle magnetic resonance heart film imaging method, device and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910625835.9A CN110338795B (en) | 2019-07-11 | 2019-07-11 | Radial golden angle magnetic resonance heart film imaging method, device and equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110338795A CN110338795A (en) | 2019-10-18 |
| CN110338795B true CN110338795B (en) | 2023-09-01 |
Family
ID=68175129
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910625835.9A Active CN110338795B (en) | 2019-07-11 | 2019-07-11 | Radial golden angle magnetic resonance heart film imaging method, device and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110338795B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110827369B (en) * | 2019-10-31 | 2023-09-26 | 上海联影智能医疗科技有限公司 | Undersampling model generation method, image reconstruction method, apparatus and storage medium |
| US11568584B2 (en) | 2019-08-26 | 2023-01-31 | Shanghai United Imaging Intelligence Co., Ltd. | Systems and methods for magnetic resonance imaging |
| US11120585B2 (en) * | 2019-11-28 | 2021-09-14 | Shanghai United Imaging Intelligence Co., Ltd. | Systems and methods for image reconstruction |
| CN111145875B (en) * | 2019-12-27 | 2023-05-12 | 上海联影医疗科技股份有限公司 | Data analysis system |
| CN111273207B (en) * | 2020-01-22 | 2022-06-10 | 上海东软医疗科技有限公司 | Magnetic resonance imaging method, magnetic resonance imaging apparatus, electronic device, and storage medium |
| US11085979B1 (en) * | 2020-02-24 | 2021-08-10 | Uih America, Inc. | System and method for magnetic resonance imaging with radial sampling of k-space |
| CN113592972B (en) * | 2021-07-30 | 2023-11-14 | 哈尔滨工业大学（深圳） | Magnetic resonance image reconstruction method and device based on multi-mode aggregation |
| CN113866695B (en) * | 2021-10-12 | 2022-08-26 | 上海交通大学 | Image acquisition and reconstruction method and system for magnetic resonance real-time guidance intervention |
| CN115067945A (en) * | 2022-08-22 | 2022-09-20 | 深圳市海清视讯科技有限公司 | Fatigue detection method, device, equipment and storage medium |
| CN117054943A (en) * | 2023-10-11 | 2023-11-14 | 山东奥新医疗科技有限公司 | Magnetic resonance imaging method, device, equipment and medium based on radial basis network |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2013159044A1 (en) * | 2012-04-19 | 2013-10-24 | New York University | System, method and computer-accessible medium for highly-accelerated dynamic magnetic resonance imaging using golden-angle radial samplng and compressed sensing |
-
2019
- 2019-07-11 CN CN201910625835.9A patent/CN110338795B/en active Active
Non-Patent Citations (1)
| Title |
|---|
| DIMENSION:Dynamic MR Imaging with Both K-space and Spatial Prior Knowledge Obtained via Multi-Supervised Network Training;Shanshan Wang,et al;《arXiv》;20181106;第1-13页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110338795A (en) | 2019-10-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110338795B (en) | Radial golden angle magnetic resonance heart film imaging method, device and equipment | |
| CN109697741B (en) | PET image reconstruction method, device, equipment and medium | |
| EP3919002A1 (en) | Ct image generation method and apparatus, computer device and computer-readable storage medium | |
| CN109598722B (en) | Image analysis method based on recurrent neural network | |
| US7558439B2 (en) | Motion artifact correction of tomographical images | |
| EP2389661B1 (en) | Nuclear image reconstruction | |
| CN104252714A (en) | Reconstruction of time-varying data | |
| US10867375B2 (en) | Forecasting images for image processing | |
| CN110751702A (en) | Image reconstruction method, system, device and storage medium | |
| CN109493394A (en) | Method, method for reconstructing and the device of magnetic resonance imaging acquisition deep learning training set | |
| CN108287324A (en) | The method for reconstructing and device of the more contrast images of magnetic resonance | |
| CN114548238A (en) | Image three-dimensional reconstruction method and device, electronic equipment and storage medium | |
| US11510642B2 (en) | Spatiotemporal reconstruction in higher dimensions of a moving vascular pulse wave from a plurality of lower dimensional angiographic projections | |
| CN114494479A (en) | System and method for simultaneous attenuation correction, scatter correction, and denoising of low dose PET images using neural networks | |
| CN111369675A (en) | Three-dimensional visual model reconstruction method and device based on lung nodule visceral layer pleural projection | |
| Novillo et al. | Unsupervised respiratory signal extraction from ungated cardiac magnetic resonance imaging at rest and during exercise | |
| CN115965837A (en) | Image reconstruction model training method, image reconstruction method and related equipment | |
| WO2020198456A1 (en) | Retrospective motion correction using a combined neural network and model-based image reconstruction of magnetic resonance data | |
| CN115908610A (en) | Method for obtaining attenuation correction coefficient image based on single-mode PET image | |
| CN115761134A (en) | Twin model generation method, system, nuclear magnetic resonance device, and medium | |
| CN113052930A (en) | Chest DR dual-energy digital subtraction image generation method | |
| US20240062332A1 (en) | System and method for improving sharpness of magnetic resonance images using a deep learning neural network | |
| US20230045406A1 (en) | System and method for hybrid imaging | |
| CN111415394B (en) | Bone hardening artifact correction method, device, computer equipment and readable storage medium | |
| Hauptmann et al. | Real-time Cardiovascular MR with Spatio-temporal De-aliasing using Deep Learning-Proof of Concept in Congenital Heart Disease |
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 |