Magnetic resonance imaging system and operation method of magnetic resonance imaging system
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- FUJIFILM CORP
- Filing Date
- 2023-06-15
- Publication Date
- 2026-07-17
Smart Images
Figure 00000000_0000_ABST
Abstract
Description
[Technical field]
[0001] The present invention relates to a magnetic resonance imaging system and a method for operating the magnetic resonance imaging system, and more particularly to a technique for improving the quality of an image reconstructed from a nuclear magnetic resonance signal. [Background technology]
[0002] One of the imaging methods using a magnetic resonance imaging (MRI) device is a diffusion weighted image (DWI).
[0003] DWI measures echo signals by applying a strong gradient magnetic field called a motion probing gradient pulse (MPG pulse) in multiple directions (axes), and images the degree of diffusion of water molecules from the difference in the effect of the MPG pulse, which is useful for diagnosing infarctions and tumors that cause diffusion restriction. Echo planar imaging (EPI), a high-speed imaging method, is often used in DWI.
[0004] There are two types of EPI: single-shot EPI and multi-shot EPI.
[0005] Single-shot EPI is a method in which multiple gradient echoes are generated by repeatedly reversing the gradient magnetic field during excitation by a single RF pulse, and k-space is filled with the data required for image reconstruction. Meanwhile, multi-shot EPI is a method in which k-space is filled with echo train data obtained by multiple shots.
[0006] Single-shot EPI has the advantage of being able to acquire data in k-space (Fourier space) where the nuclear magnetic resonance (NMR) signals are located, with a single excitation, and therefore can acquire signals at high speed. However, it also has problems such as N / 2 artifacts caused by errors in the application of the gradient magnetic field when acquiring NMR signals as echo signals while reversing the readout gradient magnetic field, and distortion and misalignment in the phase encoding direction caused by static magnetic field inhomogeneity, etc.
[0007] Compared with single-shot EPI, multi-shot EPI has the advantage that more time can be spent to collect all the data, which places less strain on the gradient magnetic field system and reduces susceptibility artifacts because there is no time for phase errors to accumulate. However, it has the disadvantage that imaging takes longer and it is more susceptible to motion artifacts.
[0008] Patent Document 1 describes a magnetic resonance imaging device that repeats a DWI sequence, collects multiple pieces of measurement data (k-space data) for the same subject, each consisting of the number of echo signals required for one image, and weights and adds the multiple pieces of measurement data (multiple images) to reconstruct one image.
[0009] Patent Document 2 describes that echo signals are acquired in three different directions by a multi-shot EPI sequence. It also describes that a complete k-space echo signal data set is acquired by four shots, and that the echo signals are acquired using signal averaging to improve the signal-to-noise ratio (SNR). Note that the number of average signals is set to 3 (NSA: number of signal average), and three signal averaging steps are used. [Prior art documents] [Patent documents]
[0010] [Patent Document 1] JP 2023-32832 A [Patent Document 2] JP 2019-503750 A Summary of the Invention [Problem to be solved by the invention]
[0011] Patent Documents 1 and 2 describe taking images multiple times and adding the resulting images (weighted addition, arithmetic averaging) in order to improve the SNR. However, adding multiple images has the problem that the SNR and G-factor cannot be sufficiently improved due to the subject's body movements with each image capture.
[0012] The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a magnetic resonance imaging system and an operating method of a magnetic resonance imaging system that can improve the image quality of a reconstructed image. [Means for solving the problem]
[0013] The invention according to a first aspect is a magnetic resonance imaging system comprising a magnetic resonance imaging device, a processor, and at least one memory, in which the processor causes the magnetic resonance imaging device to image the same slice of a subject multiple times in accordance with an imaging sequence, collects measurement data indicating nuclear magnetic resonance signals corresponding to the multiple imaging times from the magnetic resonance imaging device, obtains a common solution to be used when generating one final image in real space from the measurement data collected in the multiple imaging times, and generates a final image using the measurement data collected in the multiple imaging times and the common solution.
[0014] According to a first aspect of the present invention, when generating a single final image in real space from measurement data obtained by imaging the same slice of a subject multiple times, the signal-to-noise ratio of the final image can be improved by generating the final image using a common solution.
[0015] In the magnetic resonance imaging system according to the second aspect of the present invention, in the first aspect, the imaging sequence is preferably a multi-shot sequence or a single-shot sequence, and the multi-shot sequence or the single-shot sequence is preferably executed multiple times. Also, the multi-shot sequence or the single-shot sequence is preferably a multi-shot EPI sequence or a single-shot EPI sequence.
[0016] In the magnetic resonance imaging system according to a third aspect of the present invention, in the first or second aspect, the processor preferably generates a diffusion-weighted image as a final image.
[0017] A magnetic resonance imaging system according to a fourth aspect of the present invention is any one of the first to third aspects, wherein the magnetic resonance imaging apparatus is N ch The receiver coil is a multi-channel array coil, and the processor determines the number of shots in the imaging sequence as N sh In this case, N ch ×N sh Collect N sheets of measurement data. ch ×N sh N ch ×N sh If the number of times of imaging is N, the intermediate images for each of the N imaging times are S i (i = 1 to N), and the determinant when the final image is m is the following [Equation 1]:
[0018]
number
[0019]
number
[0020]
number
[0021] A magnetic resonance imaging system according to a fifth aspect of the present invention is the magnetic resonance imaging system of the fourth aspect, wherein the imaging sequence is a multi-shot sequence, and the number of shots in the multi-shot sequence is N sh If the speed multiplication rate of the parallel imaging method is R, the number of lines that are thinned out to 1 / R are arranged in the phase encoding direction of the k-space. The measurement data is arranged in the phase encoding direction at a rate of 1 / (N sh ×R), and N sh The matrix B is collected in i is N ch The coil sensitivity matrix of the receiving coil consisting of channels and N sh a phase shift matrix indicating a phase shift corresponding to the arrangement error in the k-space of the measurement data acquired in separate times; sh and a phase variation matrix indicating a phase variation caused by a body movement of the subject between the measurement data collected at different times, and the processor calculates the coil sensitivity matrix, the phase shift matrix, the phase variation matrix, and the N receiving coils. ch It is preferable to calculate the common solution D based on the noise correlation matrix between the channels.
[0022] A magnetic resonance imaging system according to a sixth aspect of the present invention is the magnetic resonance imaging system according to the fifth aspect, wherein the multi-shot sequence includes a main scan and a navi scan for acquiring a main echo signal and a navi echo signal as measurement data, the memory stores a coil sensitivity matrix, a phase shift matrix, and a noise correlation matrix, and the processor acquires the coil sensitivity matrix, the phase shift matrix, and the noise correlation matrix from the memory, and performs N sh It is preferable to calculate the phase variation matrix based on the Navi-Echo signals.
[0023] That is, the coil sensitivity matrix, the phase shift matrix, and the noise correlation matrix are matrices that can be measured in advance and stored in a memory, and are read out from the memory when calculating the common solution D. On the other hand, the phase variation matrix is N sh The matrix indicates a phase shift caused by the subject's body movement between the measurement data collected in separate times. sh A phase variation matrix is calculated based on a navi echo signal acquired together with a main echo signal.
[0024] A magnetic resonance imaging system according to a seventh aspect of the present invention is any one of the first to third aspects, wherein the magnetic resonance imaging apparatus is N ch The receiver coil is a multi-channel array coil, and the processor determines the number of shots in the imaging sequence as N sh In this case, N ch ×N sh Collect N sheets of measurement data. ch ×N sh N ch ×N sh If the number of times of imaging is N, the intermediate images for each of the N imaging times are S i (i = 1 to N), and the determinant when the final image is m is the following [Equation 1]:
[0025]
number
[0026]
number
[0027]
number
[0028] An eighth aspect of the present invention relates to a magnetic resonance imaging system according to the seventh aspect, wherein the imaging sequence is a multi-shot sequence, and the number of shots in the multi-shot sequence is N sh If the speed multiplication rate of the parallel imaging method is R, the number of lines that are thinned out to 1 / R are arranged in the phase encoding direction of the k-space. The measurement data is arranged in the phase encoding direction at a rate of 1 / (N sh ×R), and N sh The matrix B is collected in i is N ch The coil sensitivity matrix of the receiving coil consisting of channels and N sh a phase shift matrix indicating a phase shift corresponding to the arrangement error in the k-space of the measurement data acquired in separate times; shand a phase variation matrix indicating a phase variation caused by a body movement of the subject between the measurement data collected at different times, and the processor calculates the coil sensitivity matrix, the phase shift matrix, the phase variation matrix, and the N receiving coils. ch It is preferable to calculate the common solution D based on the noise correlation matrix between the channels.
[0029] A magnetic resonance imaging system according to a ninth aspect of the present invention is the magnetic resonance imaging system according to the eighth aspect, wherein the multi-shot sequence includes a main scan and a navi scan for acquiring a main echo signal and a navi echo signal as measurement data, the memory stores a coil sensitivity matrix, a phase shift matrix, and a noise correlation matrix, and the processor acquires the coil sensitivity matrix, the phase shift matrix, and the noise correlation matrix from the memory, and sh It is preferable to calculate the phase variation matrix based on the Navi-Echo signals.
[0030] In the magnetic resonance imaging system according to a tenth aspect of the present invention, in the fourth or seventh aspect, it is preferable that the imaging sequence is a single shot sequence to which a parallel imaging method is applied, the memory stores the common solution D, and the processor acquires the common solution D from the memory. In the case of a single shot sequence, the common solution D does not vary, so it can be calculated in advance and stored in the memory, and the processor acquires the intermediate image S all When generating the final image from the image, the common solution D can be read from the memory and used.
[0031] An eleventh aspect of the invention is a method for operating a magnetic resonance imaging system including a magnetic resonance imaging device, a processor, and at least one memory, the method including the steps of: the processor causing the magnetic resonance imaging device to image the same slice of a subject multiple times according to an imaging sequence, where N is an integer greater than or equal to 2; the processor collecting measurement data indicating nuclear magnetic resonance signals corresponding to the multiple imaging operations from the magnetic resonance imaging device; the processor acquiring a common solution to be used when generating one final image in real space from the measurement data collected in the multiple imaging operations; and the processor generating a final image using the measurement data collected in the multiple imaging operations and the common solution. Effect of the Invention
[0032] According to the present invention, a common solution is obtained that is common when generating a single final image in real space from measurement data collected by imaging the same slice of a subject multiple times, and the final image is generated using the measurement data collected by the multiple imaging operations and the common solution, thereby improving the signal-to-noise ratio of the final image and enhancing the image quality of the final image. [Brief description of the drawings]
[0033] [Figure 1] FIG. 1 is a diagram showing a schematic configuration of a magnetic resonance imaging system according to the present invention. [Diagram 2] FIG. 2 is a diagram showing an example of a multi-shot EPI sequence. [Diagram 3] FIG. 3 is a diagram showing an example of k-space data obtained by a main scan and a navi scan. [Figure 4] FIG. 4 is a functional block diagram showing functions of the processor of the control device shown in FIG. [Diagram 5] FIG. 5 is a diagram showing, in a matrix representation, the relationship between a pixel in a main scan image acquired in one imaging session of a multi-shot EPI sequence and a pixel after expansion calculated from that pixel. [Figure 6]FIG. 6 shows, in mathematical form, the difference between the conventional reconstruction method and the reconstruction method according to the present invention. [Figure 7] FIG. 7 is a diagram showing a reconstructed image and G-factors generated by a conventional reconstruction method. [Figure 8] FIG. 8 is a diagram showing a comparative example showing a reconstructed image and a G-factor map generated by a conventional reconstruction method and a reconstructed image and a G-factor map generated by the reconstruction method of the present invention. [Figure 9] FIG. 9 is a diagram showing a G-factor map of all slices obtained by the conventional reconstruction method and a G-factor map of all slices obtained by the reconstruction method of the present invention. [Figure 10] FIG. 10 is an enlarged view of a portion of the G-factor map shown in FIG. 9 according to the reconstruction method of the present invention. [Figure 11] FIG. 11 is a diagram showing the k-space in which each echo signal is arranged when the number of shots in a multi-shot EPI sequence is three. [Figure 12] FIG. 12 is a diagram showing an example of a single-shot EPI sequence. [Figure 13] FIG. 13 is a flow chart illustrating an embodiment of a method of operating a magnetic resonance imaging system according to the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0034] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, preferred embodiments of a magnetic resonance imaging system and an operating method of a magnetic resonance imaging system according to the present invention will be described with reference to the accompanying drawings.
[0035] [Magnetic resonance imaging system configuration] FIG. 1 is a diagram showing a schematic configuration of a magnetic resonance imaging system according to the present invention.
[0036] As shown in FIG. 1, the magnetic resonance imaging system 10 includes a magnetic resonance imaging apparatus (MRI apparatus) 100 and a control apparatus 200.
[0037] 1 includes a static magnetic field generating magnet 104 that generates a uniform static magnetic field in an imaging space in which a subject 102 is placed, a gradient magnetic field coil 106 that generates a gradient magnetic field in the imaging space, an RF (Radio Frequency) coil (transmitting coil) 108 that generates a high frequency magnetic field that generates a nuclear magnetic resonance signal (NMR signal) in the nuclei of atoms that constitute the tissue of the subject 102, and an RF probe (receiving coil) 110 that detects the NMR signal generated from the subject 102. The subject 102 is usually placed in the imaging space while lying on a bed 103.
[0038] A sequencer 118 sends commands to the radio frequency magnetic field generator 112 and the gradient magnetic field power supply 116 in accordance with an imaging sequence (pulse sequence), causing them to generate a radio frequency magnetic field and a gradient magnetic field, respectively. The generated radio frequency magnetic field is applied to the subject 102 as a pulsed radio frequency magnetic field (RF pulse) through a transmission coil 108. An NMR signal generated from the subject 102 is received by a receiving coil 110 and detected by a receiver 114. The NMR signal is generally collected as a gradient echo or a spin echo, and is therefore referred to as an echo signal here.
[0039] The gradient coil 106 is composed of gradient coils in three directions, X, Y, and Z, and generates gradient magnetic fields in response to signals from a gradient power supply 116. The receiving coil 110 in this example is composed of an M-channel multiple array coil, where M is an integer equal to or greater than 2.
[0040] The nuclear magnetic resonance frequency (detection reference frequency f0) used as the detection reference in the receiver 114 is set by the sequencer 118. The sequencer 118 controls each part so that it operates at pre-programmed timing and intensity. Among the programs, those that particularly describe the timing and intensity of the RF pulse, gradient magnetic field, and signal reception are called a pulse sequence.
[0041] Although various pulse sequences are known depending on the purpose, this embodiment adopts a multi-shot sequence that uses two or more excitation RF pulses to collect echo signals for image generation, and in particular, adopts a multi-shot EPI sequence that performs imaging by the EPI method. Also, an MPG pulse is used to obtain a diffusion weighted image.
[0042] FIG. 2 is a diagram showing an example of a multi-shot EPI sequence.
[0043] Figure 2(A) shows the application timing of the RF pulse, Figures 2(B), (C) and (D) show the application timing of the gradient magnetic field pulses in the slice direction, phase encoding direction and readout direction, respectively, and Figure 2(E) shows the acquisition timing of the echo signals (main echo signal, navi echo signal).
[0044] 2, in the multi-shot EPI sequence, an excitation RF pulse 301 is applied together with a slice selection gradient magnetic field 303, and then an inversion RF pulse 302 is applied together with a slice selection gradient magnetic field 304 to excite a desired slice. A high-intensity MPG pulse 309 is applied before and after the inversion RF pulse 302. Next, after a phase encoding gradient magnetic field 305 is applied, a blip-shaped phase encoding gradient magnetic field 306 and a readout gradient magnetic field 307 that inverts the polarity are applied in succession, and a main echo signal 308 is collected during the application of the inverting readout gradient magnetic field 307. In this manner, a main scan 300 that collects the main echo signal 308 is performed.
[0045] After the main scan 300, a navi scan 310 is performed.
[0046] In the navi scan 310, the polarity of the phase encoding gradient magnetic field 316 is set to the same polarity as the phase encoding gradient magnetic field 306 of the main scan 300, and a navi echo signal 318 is collected each time the readout 317 is inverted. The number and range of the navi echo signals 318 to be collected are not particularly limited as long as an image with the same FOV (Field of View) as the main scan image is obtained as the reconstructed navi scan image, but the applied strength of the phase encoding pulse 306 is set to be smaller than that of the main scan 300, and data near the center of k-space (low frequency region) is collected.
[0047] In the multi-shot EPI sequence, excitation RF pulses are repeatedly generated at time intervals of TR (time to repeat), and a main echo signal 308 and a navi echo signal 318 are collected and arranged in the k-space.
[0048] FIG. 3 is a diagram showing an example of k-space data obtained by a main scan and a navi scan.
[0049] In this example, the main scan 300 and the navi scan 310 collect echo signals using the same polarity of the phase encoding gradient magnetic field, so that in the main scan and the navi scan, the k-space is scanned from top to bottom, as shown in Figures 3(A) and (B), respectively.
[0050] In a multi-shot EPI sequence, main echo signals are acquired in such a way that the phase encoding direction of the main echo signals acquired for each shot is shifted (so that the signals are positioned in different regions in the k-space).
[0051] The k-space shown in FIG. 3(A) is shown for a case where the number of shots in a multi-shot EPI sequence is three, and the main echo signals acquired for each shot are arranged on different trajectories in the k-space.
[0052] The number of shots (number of generated echoes) in a multi-shot EPI sequence is about 2 to 10, and the TR is about 1000 ms to 7000 ms. To improve spatial resolution, diffusion-weighted imaging with a multi-shot sequence is advantageous, as it divides and acquires the echo signals required for one image.
[0053] Since the SNR of the diffusion weighted image (DWI) obtained by one multi-shot EPI sequence and reconstructed is low, the same slice of the subject is imaged multiple times. In addition, the MGP pulse is applied in each of three orthogonal axes (1st axis, 2nd axis, 3rd axis) to obtain the DWI corresponding to each axis.
[0054] 1, the control device 200 controls the operation of the MRI device 100 via the sequencer 118 described above, and also receives signals detected by the MRI device 100 (receiver 114) 10 (main echo signal from main scan and navi echo signal from navi scan), and performs various signal processing such as image reconstruction. The receiver 114 performs quadrature phase detection of the echo signal, which is an analog wave, using a set detection reference frequency f0, converts it into raw data of complex numbers consisting of real and imaginary parts, and then transmits it to the control device 200. In this example, this raw data is also called echo signal or measurement data.
[0055] The control device 200 can be configured using a computer. The computer applied to the control device 200 may be a personal computer or a workstation.
[0056] The control device 200 includes, as hardware, a processor 202, a memory 204, an input / output interface 206, a display 208, an operation unit 210, and the like.
[0057] The processor 202 is composed of a CPU (Central Processing Unit) etc., and controls each part of the control device 200 and the MRI device 100 in an integrated manner, and executes various programs stored in the memory 204 to realize various functions described below.
[0058] The memory 204 is one or more memories including a flash memory, a read-only memory (ROM), a random access memory (RAM), a hard disk drive, etc. The flash memory, the ROM, and the hard disk drive are non-volatile memories that store an operation system, a program that causes the processor 202 to function as the control device 20, various pulse sequences, calculation formulas and parameters used for image reconstruction, reconstructed MRI images, etc.
[0059] The RAM functions as a working area for processing by the processor 202, and temporarily stores programs and the like stored in the non-volatile memory. The RAM also functions as a place (k-space) for temporarily storing echo signals (raw data). Note that the processor 202 may incorporate a part of the memory 204 (RAM).
[0060] The input / output interface 206 includes a communication unit connectable to a network, a connection unit connectable to an external device, etc. As the connection unit connectable to an external device, a Universal Serial Bus (USB), a High-Definition Multimedia Interface (HDMI) (HDMI is a registered trademark), etc. can be applied.
[0061] The processor 202 communicates with a device (the sequencer 118 in this example) disposed in the MRI apparatus 100 via the input / output interface 206, thereby transmitting and receiving necessary information. Note that a part of the sequencer 118 may be provided on the control device 200 side.
[0062] The display 208 is used as a part of a GUI (Graphical User Interface) when receiving an MRI image and an input from an operation unit 210 .
[0063] The operation unit 210 includes a mouse, a keyboard, etc., and functions as a part of a GUI that uses the display operation window of the display 208 and receives input from an operator.
[0064] That is, the operation unit 210 and the display 208 function as a GUI for the operator to start and stop (pause) the MRI apparatus 100, select a pulse sequence, and input imaging conditions, processing conditions, and the like.
[0065] FIG. 4 is a functional block diagram showing functions of the processor of the control device shown in FIG.
[0066] As shown in FIG. 4, the processor 202 executes various programs stored in the memory 204 to function as a measurement control unit 220, an image calculation unit 230, and a display control unit 240.
[0067] The measurement control unit 220 includes a main scan control unit 222 and a navi scan control unit 224, and sends a predetermined pulse sequence and imaging conditions (imaging parameters) designated by an operator to the sequencer 118 (FIG. 1) of the MRI apparatus 100, and controls the MRI apparatus 100 via the sequencer 118. In this example, the multi-shot EPI sequence shown in FIG. 2 etc. is selected as the predetermined pulse sequence.
[0068] The main scan control unit 222 and the navi scan control unit 224 send commands to the sequencer 118 according to the set conditions. Under the control of the sequencer 118, the MRI apparatus 100 executes a multi-shot EPI sequence, acquires main echo signals by the main scan and navi echo signals by the navi scan, and arranges the main echo signals and navi echo signals in the corresponding k-spaces.
[0069] The image calculation unit 230 includes an inverse Fourier transform unit 232, a phase variation matrix calculation unit 234, a common solution calculation unit 236, an MRI image generation unit 238, and the like.
[0070] The inverse Fourier transform unit 232 is a part that performs inverse Fourier transform on echo signals in the spatial frequency domain arranged in the k-space and converts them into an image in real space.
[0071] One of the challenges of DWI generated by multi-shot EPI is that the phase differs from shot to shot (causing phase disturbance) in moving parts of the subject due to the influence of a pair of large gradient magnetic fields (MPG) applied for each shot.
[0072] The phase variation matrix calculation unit 234 is a part that calculates a phase variation matrix based on the Navi echo signal of each shot, and calculates information (phase variation matrix) indicating image variation between shots based on the Navi scan image reconstructed from the Navi echo signal. This phase variation matrix is used to correct distortion between images reconstructed from the main echo signal of each shot, which is caused by the subject's body movement between shots.
[0073] Since DWI has a low SNR, the same slice of the subject is imaged multiple times. Conventionally, multiple intermediate images are generated by multiple imaging, and the SNR of the final image (DWI) is improved by averaging the multiple intermediate images. In addition, to generate each intermediate image, an individual correction solution (matrix) is calculated, and the corresponding intermediate image is corrected using the calculated individual solution.
[0074] The common solution calculation unit 236 is a part that calculates a common solution used when generating one final image (DWI) in real space from measurement data (echo signals) collected corresponding to multiple imaging operations, without calculating individual correction solutions as in the conventional technology.
[0075] The MRI image generating unit 238 is a part that uses echo signals acquired by multiple imaging sessions and the common solution calculated by the common solution calculating unit 236 to generate one final image (DWI).
[0076] The details of the calculation processes performed by the common solution calculation unit 236 and the MRI image generation unit 238 will be described later.
[0077] Another issue with DWI is that images are prone to distortion and blurring, due to the fact that EPI is easily affected by static magnetic field inhomogeneity and that the direction of the readout gradient magnetic field changes for each echo, which is a problem unique to EPI.
[0078] The image calculation unit 230 includes a displacement amount calculation unit and a distortion correction unit, not shown. Because the main echo signal from the main scan and the navi echo signal from the navi scan have the same phase encoding application polarity, distortions occurring in the main scan image and the navi scan image due to accumulation of phase errors caused by static magnetic field inhomogeneity, etc., are distortions in the same direction. The displacement amount calculation unit uses the navi scan image to calculate the amount of displacement (amount of displacement in real space) used for distortion correction. The distortion correction unit 223 uses this displacement amount to perform distortion correction on the main scan image.
[0079] The display control unit 240 controls the display device 208 (FIG. 1) to display the DWI calculated by the image calculation unit 230, or the MRI image captured and reconstructed by a pulse sequence other than the multi-shot EPI sequence, and necessary additional information (e.g., imaging conditions, information on the subject, etc.). The processor 202 also controls the memory 204 to store the DWI with the additional information.
[0080] The above-mentioned functions of the processor 202 are realized by a CPU or a GPU executing a pre-designed program. Note that some of the functions of the processor 202 may be realized by hardware such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).
[0081] [Principle of the present invention] Next, the principle of the present invention related to the calculation processing by the common solution calculation unit 236 and the MRI image generation unit 238 shown in FIG. 4 will be described.
[0082] In the example shown below, the number of shots (N sh ) twice.
[0083] In this example, the parallel imaging method is used. The parallel imaging method is a technology that uses multiple coils (multiple array coils) and enables imaging in a short time by thinning out the phase encoding. Parallel imaging includes the SENSE (sensitivity encoding) system, which performs unfolding processing in real space, and the SMASH (Simultaneous Acquisition of Spatial Harmonics) system, which performs processing in k-space. In this example, the SENSE system is used as the parallel imaging, and the reduction factor R, which indicates the thinning rate of the parallel imaging method, is set to 2.
[0084] Number of shots in a multi-shot EPI sequence, N sh By setting the speed ratio R of the parallel imaging method to 2, the measurement data thinned out by half in the phase encoding direction of k-space is thinned out by a quarter in the phase encoding direction (1 / (N sh ×R)) and thinned out twice (N sh ) are collected.
[0085] The receiving coil 110 in this example has N ch It is composed of a multiple array coil with six channels (N ch The system is composed of multiple array coils (=6).
[0086] Number of shots in a multi-shot EPI sequence, N sh When one multi-shot EPI sequence is executed twice (sh1, sh2), 12 images (i.e., N ch (N ch = 6) × Number of shots N sh (N sh Echo signals corresponding to the image (=2) are acquired.
[0087] Then, by performing an inverse Fourier transform on the echo signals corresponding to the 12 images, 12 images in real space can be generated.
[0088] The echo signals collected by each channel coil are thinned out to one-fourth in the phase encoding direction of k-space, thereby shortening the scan time and enabling faster imaging per image.
[0089] On the other hand, the FOV of each image becomes smaller and aliasing occurs. In this example, the echo signals corresponding to each image are thinned out to one-fourth in the phase encoding direction of the k-space, so the image generated by the inverse Fourier transform becomes an aliased image (aliased image), and the FOV becomes one-fourth. Now, if the size of the FOV without aliasing is 256 pixels x 256 pixels, the size of the FOV of the aliased image is 256 pixels x 64 pixels.
[0090] For the aliased image, a sensitivity map for each coil of the receiving coil 110 is used, and an unfolding process is performed to obtain an image without aliasing.
[0091] FIG. 5 is a diagram showing, in a matrix representation, the relationship between a pixel in a main scan image acquired in one imaging session of a multi-shot EPI sequence and a pixel after expansion calculated from that pixel.
[0092] In Fig. 5, the left side of the equation expressed as a matrix is the 12 acquired signals (s1-s 6, s1-s6), and the right side represents the coil sensitivity matrix C, the phase shift matrix A, the phase variation matrix P, and the acquired signals (s1-s 6, s1-s6) corresponding to four pixels of signal m (m A -m D ), where A to D in the superscript of the signal m represent the number of shots N sh × speed ratio R, in this example, 4 (=N sh ×R).
[0093] Acquired signal (s1-s 6, s1-s6) are the six channels (N ch =6) × 2 shots (N sh The signals from pixels at the same positions in 12 folded images (256 pixels x 64 pixels) obtained by inverse Fourier transform of the echo signals acquired by the inverse Fourier transform (Fourier transform = 2) are shown.
[0094] The coil sensitivity matrix C is a sensitivity map (c A -c1 D ,c2 A -c2 D , …, c6 A -c6 D ) is a matrix showing the sensitivity map (c1 A -c D ~c6 A -c6 D ) can be measured in advance by executing a reference scan and stored in the memory 204.
[0095] This sensitivity map (c1 A -c D ~c6 A -c6 D ) is used to unfold the acquired folded images (12 folded images in this example) corresponding to each coil to generate a single image without folding. In this example, 12 redundant acquired signals (s1-s 6, s1-s6) and the sensitivity map with different sensitivities for each coil (c1 A -c D ~c6 A -c6 D ) and based on the acquired signal (s1-s 6, s1-s6) corresponding to the four pixels of the signal m (m A -m D ) can be calculated.
[0096] Then, the acquired signal (s1-s 6, s1-s6), the signal m (m A -mD ) can generate a single image (256 pixels x 256 pixels) without aliasing.
[0097] The measurement data was thinned out by half in the phase encoding direction of k-space twice (N sh ), which are collected separately using the phase shift matrix A(a A ~a D ) is a matrix corresponding to the k-space placement error of the two sets of measurement data that fill the k-space. The phase shift matrix A can be calculated based on a navi-scan image reconstructed from navi-echo signals that correspond to the two sets of measurement data (main echo signals) that fill the k-space.
[0098] In FIG. 5, the phase variation matrix P(p A ~p D ) twice (N sh ), which is a matrix that indicates the phase fluctuation caused by the subject's body movement between the measurement data (echo signals) collected in two separate steps. Like the phase shift matrix A, the phase fluctuation matrix P can be calculated based on the navi-scan image reconstructed from navi-echo signals corresponding to the two main echo signals that fill the k-space. Since the subject's body movement is irregular, the phase fluctuation matrix P is calculated based on the navi-scan image reconstructed from navi-echo signals corresponding to the two main echo signals that fill the k-space. sh ) and is calculated by the phase variation matrix calculation unit 234 each time measurement data (echo signals) is acquired.
[0099] Now, assuming that the coil sensitivity matrix C, the phase shift matrix A, and the phase variation matrix P are matrix B=CAP, the intermediate image S of the matrix representation shown in FIG. 5 is expressed by the following determinant:
[0100]
number
[0101] (where i is a parameter indicating the imaging number of multiple imaging sessions) It can be expressed as:
[0102] In the conventional MRI image reconstruction method, a solution to the equation [1] is calculated for each imaging by a multi-shot EPI sequence, and an image m with aliasing and the like removed is calculated by the following equation.
[0103]
number
[0104] In the formula [2], Σ represents a noise correlation matrix between multiple channels of the receiving coil.
[0105] In addition, in the formula [2], m and S are as follows: m=[m k ]: an image (column vector) with aliasing etc. removed, and k is the pixel number.
[0106] S=[S ij ]:N ch ×N sh where i is the coil number in shots sh1 and sh2, and j is the pixel number.
[0107] Fig. 6 shows the difference between the conventional reconstruction method and the reconstruction method according to the present invention in the form of mathematical expressions. In Fig. 6, the same slice of the subject is executed multiple times (four times in this example) by the MRI apparatus 100 according to the multi-shot EPI sequence.
[0108] The left side of FIG. 6 shows the above-mentioned conventional reconstruction method. If the parameter i (i=1 to 4) indicates each image taken by the four multi-shot EPI sequences, then m i to the following equation:
[0109]
number
[0110] The calculated image m i to the following equation:
[0111]
number
[0112] In contrast, the reconstruction method according to the present invention can obtain 12 images (N ch ×N sh ) folded image (intermediate image) S i (i=1 to 4), but unlike the conventional technique, each intermediate image S i Without solving for all intermediate images S1 to S N It differs from conventional reconstruction methods in that it calculates one common solution for (N=4).
[0113] That is, if the common solution is D, the common solution D can be expressed as the following [Equation 5]:
[0114]
number
[0115]
number
[0116] The common solution calculation unit 236 shown in FIG. 4 calculates the coil sensitivity matrix C, the phase shift matrix A, the noise correlation matrix Σ stored in the memory 204, and the phase variation matrix P calculated by the phase variation matrix calculation unit 234, and calculates the intermediate images S1 to S N Based on this, a common solution (matrix D) shown in [Equation 5] is calculated.
[0117] Then, the MRI image generating unit 238 shown in FIG. 4 calculates the matrix D and the matrix S allFrom this, one final image (MRI image) m is generated by equation (6).
[0118] In the above embodiment, when calculating the common solution D as shown in [Formula 5], the inverse matrix (Σ) of the noise correlation matrix (Σ) between multiple channels of the receiving coil is used. -1 ) is used, but it is not limited to this. Instead of using the noise correlation matrix (Σ), the following equation is used:
[0119]
number
[0120] <Comparative Example> Next, a comparative example will be described in which an MRI image generated by a conventional reconstruction method is compared with an MRI image generated by the reconstruction method according to the present invention.
[0121] FIG. 7 is a diagram showing a reconstructed image (MRI image) generated by a conventional reconstruction method, G-factors, etc.
[0122] In the example shown in FIG. 7, the number of shots in the multi-shot EPI sequence is four, and the number of imaging sessions in the multi-shot EPI sequence is two.
[0123] 7 are reconstructed images corresponding to three axes in which the application directions of the MPG pulse are different from each other. In addition, the three reconstructed images MPG1, MPG2, and MPG3 are images of the sixth slice of the subject.
[0124] The number of times imaging was performed using the multi-shot EPI sequence was two (NSA=2), and the number of shots in each multi-shot EPI sequence was four. 4 shots × 2NSA imaging was performed for each of the reconstructed images MPG1, MPG2, and MPG3, and eight NaviScan images corresponding to the 4 shots × 2NSA and a map showing the G-factors of two images of the 2NSA are shown.
[0125] Here, the SNR of the separated image (separated image) made up of signals separated by the parallel imaging method is known to be inversely proportional to an index called the G-factor (Geometry-factor). When the difference in sensitivity between the receiving coils at the position of the spatially overlapping signals is small, the G-factor increases and the SNR decreases. The G-factor can be considered an index for evaluating image quality.
[0126] The G-factor can be calculated using the following formula:
[0127]
number
[0128] In Figure 7, it can be seen that the G-factor corresponding to the reconstructed image MPG3 on the third axis is increased. This is thought to be because the application direction of the MPG pulse on the third axis is the same as or close to the subject's body movement direction, resulting in a decrease in SNR.
[0129] FIG. 8 is a comparative example showing a reconstructed image and a G-factor map generated by a conventional reconstruction method and a reconstructed image and a G-factor map generated by the reconstruction method of the present invention.
[0130] In the example shown in FIG. 8, imaging is performed four times using the multi-shot EPI sequence.
[0131] FIG. 8 shows four images and a G-factor map of each conventional NSA, one reconstructed image obtained by averaging the four images, and a reconstructed image and G-factor map generated based on the common solution of the present invention.
[0132] The maximum value of the G-factor map obtained by the conventional reconstruction method was 9.3, whereas the maximum value of the G-factor map obtained by the reconstruction method of the present invention was 1.5.
[0133] Thus, the G-factor obtained by the reconstruction method of the present invention is smaller than the G-factor obtained by the conventional reconstruction method, and it can be said that the SNR is improved.
[0134] FIG. 9 is a diagram showing a G-factor map of all slices obtained by the conventional reconstruction method and a G-factor map of all slices obtained by the reconstruction method of the present invention.
[0135] Figure 9 shows G-factor maps of three axes (1st axis, 2nd axis, 3rd axis) with different MPG pulse application directions for all slices (18 slices) of the subject. In the example shown in Figure 9, the number of imaging sessions using the multi-shot EPI sequence is four, as in the case of Figure 8.
[0136] Therefore, the G-factor map using the conventional reconstruction method shown in Figure 9 is 216 images (= 18 slices x 4 shots x 3 axes), while the G-factor map using the reconstruction method of the present invention is 54 images (= 18 slices x 1 x 3 axes).
[0137] The G-factor map produced by the conventional reconstruction method shown in Figure 9 shows a larger value on the third axis than on the first and second axes, but the G-factor map produced by the reconstruction method of the present invention shows lower values on all three axes.
[0138] That is, the application direction of the MPG pulse on the third axis is considered to be the same as or close to the subject's body movement direction. In the conventional reconstruction method, blurring and artifacts in the reconstructed image on the third axis are not sufficiently removed, resulting in a low SNR. In contrast, the reconstruction method of the present invention removes blurring and artifacts in the reconstructed image on the third axis, and it can be seen that the SNR is high, similar to the first and S2 axes.
[0139] FIG. 10 is an enlarged view of a portion (slice 6 to 13) of the G-factor map obtained by the reconstruction method of the present invention shown in FIG.
[0140] In the reconstruction method of the present invention, the condition (=G-factor) is improved by increasing the number of equations relative to the number of solutions.
[0141] Figure 11 shows the number of shots in a multi-shot EPI sequence, N sh FIG. 13 is a diagram showing the k-space in which each echo signal is arranged when the number of echoes is three.
[0142] In contrast to the case where there is no phase variation in the three shots, there are cases where the same phase variation occurs between the shots, and cases where the phase variation occurs randomly between the shots. The phase variation can be considered as the variation of the trajectory in the k section, and the reconstruction method of the present invention is also considered to improve the coverage rate of the k space.
[0143] The present invention is not limited to a multi-shot sequence, but can also be applied to a single-shot sequence.
[0144] FIG. 12 is a diagram showing an example of a single-shot EPI sequence.
[0145] Figure 12(A) shows the application timing of an RF pulse, Figures 12(B), (C) and (D) show the application timing of gradient magnetic field pulses in the slice direction, phase encoding direction and readout direction, respectively, and Figure 12(E) shows the acquisition timing of an echo signal.
[0146] In order to improve the SNR of DWI imaged and reconstructed by a single-shot EPI sequence, the single-shot EPI sequence is executed multiple times to image the same slice of the subject multiple times.
[0147] 12, in the single-shot EPI sequence, an excitation RF pulse 301 is applied together with a slice selection gradient magnetic field 303, and then an inversion RF pulse 302 is applied together with a slice selection gradient magnetic field 304 to excite a desired slice. A high-intensity MPG pulse 309 is applied before and after the inversion RF pulse 302. Next, a phase encoding gradient magnetic field 305 is applied, and then a blip-shaped phase encoding gradient magnetic field 306 and a readout gradient magnetic field 307 that inverts the polarity are applied in succession, and a main echo signal 308 is collected during the application of the inverting readout gradient magnetic field 307.
[0148] Note that a navi echo signal may be collected separately from the main echo signal 308. In this case, the navi echo signal can be used to correct distortion of a reconstructed image reconstructed from the main echo signal 308.
[0149] The single-shot EPI sequence in this example is a single-shot sequence that applies the parallel imaging method. That is, the speed factor R of the parallel imaging method is set, and the measurement lines thinned out to 1 / R in the phase encoding direction of the k-space are thinned out in the phase encoding direction according to the speed factor R, and collected in one measurement.
[0150] Now, if the speed multiplication rate R of the parallel imaging method applied to the single-shot EPI sequence is 3 (R=3), the measurement data thinned out to 1 / R in the phase encoding direction of k-space is thinned out to one-third in the phase encoding direction of k-space, collected in one measurement, and arranged in k-space.
[0151] In the case of a single-shot EPI sequence, since there is no k-space arrangement error compared to a multi-shot EPI sequence, calculation of the phase shift matrix A can be omitted. Also, since there is no influence of phase fluctuation caused by the subject's body movement between shots, calculation of the phase fluctuation matrix P can be omitted. Therefore, the aforementioned matrix B can be made to be only the coil sensitivity matrix C (B=C).
[0152] The coil sensitivity matrix C can be measured in advance by executing a reference scan, and the phase shift matrix A can also be commonly used after it has been measured. Therefore, the common solution D shown in [Equation 5] or [Equation 7] can be calculated in advance and stored in the memory 204.
[0153] In the case of a single-shot EPI sequence, the intermediate images S all and the common solution D (eg, matrix D read from memory 204), a final image can be generated.
[0154] [Method of operating a magnetic resonance imaging system] FIG. 13 is a flow chart illustrating an embodiment of a method of operating a magnetic resonance imaging system according to the present invention.
[0155] The processing of each step of the method of operating the magnetic resonance imaging system shown in FIG. 13 corresponds to the processing performed by the processor 202 of the control device 200 shown in FIG.
[0156] 13, the processor 202 executes a multi-shot EPI sequence (step S10). sh (N sh N: an integer of 2 or more; R: the speed factor of the parallel imaging method (R: a real number greater than 1), sh × R in the phase encoding direction of k-space, 1 / (N sh ×R), N sh Measurement data is collected from the MRI apparatus 100 in separate times (step S12). In this example, a set of a main echo signal and a navi echo signal is acquired.
[0157] The processor 202 calculates a phase variation matrix P that indicates phase variations between each piece of measurement data caused by the body movement of the subject, based on the information of the navi-echo signal (step S14).
[0158] If the number of imaging times using the multi-shot EPI sequence is N (N: an integer equal to or greater than 2), it is determined whether or not N imaging times have been completed, and the processes from step S10 to step S16 are repeated until N imaging times have been completed (step S16).
[0159] Next, the processor 202 calculates a matrix B (=CAP) corresponding to N imaging operations when the coil sensitivity matrix C, the phase shift matrix A, and the phase variation matrix P calculated in step S14 are used as the matrix B (=CAP). all (=B1~B N ) is calculated (step S18). Note that the coil sensitivity matrix C and the phase shift matrix A may be measured in advance by a reference scan or the like and stored in the memory 204 and used.
[0160] The processor 202 performs an inverse Fourier transform on the measurement data to generate all intermediate images S all (=S1~S N ) is calculated (step S20).
[0161] The processor 202 then extracts all the intermediate images S all A common solution (matrix D) used when generating one final image m from is calculated by the formula [5] or the formula [7] (step S22).
[0162] Finally, the processor 202 divides one final image m into all intermediate images S all and matrix D according to the following formula (6) (step S24).
[0163] [others] The imaging sequence applied to the present invention is not limited to a multi-shot EPI sequence or a single-shot EPI sequence, but various modifications are possible, such as applying various sequences and combining known techniques for image reconstruction.
[0164] In this embodiment, the hardware structure of a processing unit that executes various processes, such as a CPU, is various processors as shown below. The various processors include a CPU (Central Processing Unit), which is a general-purpose processor that executes software (programs) and functions as various processing units, a programmable logic device (PLD), such as an FPGA (Field Programmable Gate Array), whose circuit configuration can be changed after manufacture, and a dedicated electric circuit, such as an ASIC (Application Specific Integrated Circuit), which is a processor having a circuit configuration designed specifically for executing specific processes.
[0165] One processing unit may be configured with one of these various processors, or may be configured with two or more processors of the same or different types (for example, multiple FPGAs, or a combination of a CPU and an FPGA). Also, multiple processing units may be configured with one processor. As an example of configuring multiple processing units with one processor, first, as represented by a computer such as a client or server, there is a form in which one processor is configured with a combination of one or more CPUs and software, and this processor functions as multiple processing units. Second, as represented by a system on chip (SoC), there is a form in which a processor is used that realizes the functions of the entire system including multiple processing units with one IC (Integrated Circuit) chip. In this way, the various processing units are configured using one or more of the above various processors as a hardware structure.
[0166] Moreover, the hardware structure of these various processors is, more specifically, an electric circuit that combines circuit elements such as semiconductor elements.
[0167] Furthermore, the present invention is not limited to the above-described embodiment, and various modifications are possible without departing from the spirit of the present invention. [Explanation of symbols]
[0168] 10. Magnetic Resonance Imaging System 100 MRI machine 102 Subject 104 Static magnetic field generating magnet 106 Gradient magnetic field coil 108 Transmitting Coil 110 Receiver coil 112 High Frequency Magnetic Field Generator 114 Receiver 116 Gradient magnetic field power supply 118 Sequencer 200 Control device 202 Processor 204 Memory 206 Input / Output Interface 208 Display 210 Operation section
Claims
1. 1. A magnetic resonance imaging system comprising a magnetic resonance imaging device, a processor, and at least one memory, The processor, imaging the same slice of the subject a plurality of times with the magnetic resonance imaging apparatus according to an imaging sequence; collecting measurement data indicating nuclear magnetic resonance signals from the magnetic resonance imaging apparatus in response to the multiple imaging operations; obtaining a common solution to be used when generating one final image of the real space from the measurement data collected by the multiple imaging operations; generating the final image using the metrology data collected from the multiple captures and the common solution; Magnetic resonance imaging system.
2. The imaging sequence is a multi-shot sequence or a single-shot sequence, and the multi-shot sequence or the single-shot sequence is executed a plurality of times.
10. The magnetic resonance imaging system of claim 1.
3. The processor generates a diffusion weighted image as the final image.
3. The magnetic resonance imaging system of claim 2.
4. The magnetic resonance imaging apparatus includes: ch A receiving coil including a multiple array coil of channels, The processor determines the number of shots in the imaging sequence as N sh In this case, N ch ×N sh Collecting sheets of said measurement data, The N ch ×N sh N pieces of the measurement data are obtained by inverse Fourier transform. ch ×N sh The intermediate images are images of the real space, and when the number of times of imaging is N, the intermediate images for each of the N imaging times are S i (i=1 to N), and the final image is m. The determinant is expressed by the following [Equation 1]: [0010] (However, B i : Matrix B including the coil sensitivity matrix i (i = 1 to N) When the common solution calculated by the formula 1 is represented as D, the common solution D is expressed by the following formula 2: [0025] (However, B all : The matrix B i (i = 1 to N), Σ: N of the receiving coil ch Inter-channel noise correlation matrix) and The processor, The N ch ×N sh The measurement data is inverse Fourier transformed to obtain the N ch ×N sh The intermediate images S i Generate The intermediate images S generated for each of the N imaging times i All the intermediate images of S all Then, the intermediate image S all Based on the common solution D, the following [Equation 3] is calculated: [0030] generating the final image m by 10. The magnetic resonance imaging system of claim 1.
5. the imaging sequence is a multi-shot sequence, The number of shots in the multi-shot sequence is N sh When the number of lines is thinned out to 1 / R in the phase encoding direction of the k-space, the measurement data is arranged in a manner such that the number of lines is 1 / (N sh ×R), and N sh It was collected in batches, The matrix B i is the N ch a coil sensitivity matrix of the receiving coil consisting of channels; sh a phase shift matrix indicating a phase shift corresponding to a position error in the k-space of the measurement data acquired in separate times; sh a phase variation matrix indicating a phase variation caused by a body movement of a subject between the measurement data collected in separate times, The processor calculates the coil sensitivity matrix, the phase shift matrix, the phase variation matrix, and the N number of receive coils. ch Calculating the common solution D based on a noise correlation matrix between channels; 5. A magnetic resonance imaging system according to claim 4.
6. the multi-shot sequence includes a main scan and a navi scan for acquiring a main echo signal and a navi echo signal as the measurement data, the memory stores the coil sensitivity matrix, the phase shift matrix, and the noise correlation matrix; The processor retrieves the coil sensitivity matrix, the phase shift matrix, and the noise correlation matrix from the memory, and sh calculating the phase variation matrix based on the navi-echo signals; 6. A magnetic resonance imaging system as claimed in claim 5.
7. The magnetic resonance imaging apparatus includes: ch A receiving coil including a multiple array coil of channels, The processor determines the number of shots in the imaging sequence as N sh In this case, N ch ×N sh Collecting sheets of said measurement data, The N ch ×N sh N pieces of the measurement data are obtained by inverse Fourier transform. ch ×N sh The intermediate images are images of the real space, and when the number of times of imaging is N, the intermediate images for each of the N imaging times are S i (i=1 to N), and the final image is m. The determinant is expressed by the following [Equation 1]: [0010] (However, B i : Matrix B including the coil sensitivity matrix i (i = 1 to N) When the common solution calculated by the formula 1 is represented as D, the common solution D is expressed by the following formula 2: [0025] (However, B all : The matrix B i (i = 1 to N) and The processor, The N ch ×N sh The measurement data is inverse Fourier transformed to obtain the N ch ×N sh The intermediate images S i Generate The intermediate images S generated for each of the N imaging times i All the intermediate images of S all Then, the intermediate image S all Based on the common solution D, the following [Equation 3] is calculated: [0030] generating the final image m by 10. The magnetic resonance imaging system of claim 1.
8. the imaging sequence is a multi-shot sequence, The number of shots in the multi-shot sequence is N sh When the number of lines is thinned out to 1 / R in the phase encoding direction of the k-space, the measurement data is arranged in a manner such that the number of lines is 1 / (N sh ×R), and N sh It was collected in batches, The matrix B i is the N ch a coil sensitivity matrix of the receiving coil consisting of channels; sh a phase shift matrix indicating a phase shift corresponding to a position error in the k-space of the measurement data acquired in separate times; sh a phase variation matrix indicating a phase variation caused by a body movement of a subject between the measurement data collected in separate times, The processor calculates the coil sensitivity matrix, the phase shift matrix, the phase variation matrix, and the N number of receive coils. ch Calculating the common solution D based on a noise correlation matrix between channels; 8. A magnetic resonance imaging system according to claim 7.
9. the multi-shot sequence includes a main scan and a navi scan for acquiring a main echo signal and a navi echo signal as the measurement data, the memory stores the coil sensitivity matrix, the phase shift matrix, and the noise correlation matrix; The processor retrieves the coil sensitivity matrix, the phase shift matrix, and the noise correlation matrix from the memory, and sh calculating the phase variation matrix based on the navi-echo signals; 9. A magnetic resonance imaging system according to claim 8.
10. the imaging sequence is a single-shot sequence to which a parallel imaging method is applied, The memory stores the common solution D; The processor retrieves the common solution D from the memory.
8. A magnetic resonance imaging system according to claim 4 or 7.
11. 1. A method of operating a magnetic resonance imaging system comprising a magnetic resonance imaging device, a processor, and at least one memory, comprising the steps of: the processor causes the magnetic resonance imaging apparatus to image the same slice of the subject multiple times according to an imaging sequence, where N is an integer equal to or greater than 2; The processor collects measurement data indicating a nuclear magnetic resonance signal from the magnetic resonance imaging device corresponding to the multiple imaging operations; obtaining a common solution used by the processor to generate a single final image of real space from the measurement data collected in the multiple imaging operations; generating the final image using the metrology data collected in the multiple captures and the common solution; 4. A method of operating a magnetic resonance imaging system comprising: