A method of cartesian sampled dual-nuclear simultaneous magnetic resonance imaging and image reconstruction

Abstract

This invention discloses a Cartesian sampling dual-core synchronous magnetic resonance imaging and image reconstruction method, with sequential excitation... 1 For H and X nuclei, an adjustment parameter ap1 is introduced to adjust the X nuclei layer thickness, and a pre-dispersive gradient pulse is applied in advance along the layer direction. Adjustment parameters ap2 and ap3 are introduced to adjust the X nuclei frequency and phase encoding direction field of view. The X nuclei are spatially encoded and their echo1 is acquired. Frequency and phase encoding compensation gradients are applied to allow the proton k-trajectory to evolve to the target point before frequency encoding and acquisition. 1 H's echo2, apply the frequency coding gradient in the opposite direction and acquire 1 H's echo3, X's TE1, 1 The independent control of TE2 and TE3 of H, along with their respective flip angles, enables adjustment of the three-directional resolution of the X-core, facilitating the acquisition of multi-parameter weighted values. 1 H-image. This invention solves the problem of X-nucleus and... 1 The issue of significant differences in H-image resolution can be addressed by adjusting the X-core resolution according to the specific application scenario.

Description

Technical Field

[0001] This invention relates to the field of magnetic resonance imaging technology, and more specifically to a Cartesian sampling dual-core synchronous magnetic resonance imaging and image reconstruction method. Background Technology

[0002] Nuclear magnetic resonance imaging (MRI) technology manipulates atomic nucleus spin through flexible pulse sequences and adjustable parameters to obtain images with different parameter weights, and has wide applications in life sciences and other fields. Traditional magnetic resonance imaging (MRI) is used to visualize protons (…) within living organisms. 1 High-field MRI (H) imaging provides morphological and indirect functional information, offering advantages such as imaging in any cross-sectional direction, high soft tissue contrast, rich image information, and no ionizing radiation. With the development of high-field MRI systems, the imaging of endogenous non-proton nuclei (NPNs) related to physiological changes within organisms has become increasingly important. 23 Na、 31 P, 35 / 37 Cl、 17 O, etc.) and the nucleus in exogenous probes ( 19 F, 129 Xe, 3 He 13 Effective imaging and spectroscopic detection (such as C) are also gradually developing. Detection of non-proton nuclides can provide additional information that proton detection cannot obtain, helping to explain subcellular or molecular events such as cell function, ion disorders, energy metabolism, tissue pH, and drug targeting sites, and contributing to in-depth research on pathological processes.

[0003] Imaging of nonproton nuclei usually requires... 1 The combined use of h-MRI is primarily because nonproton nuclei are locally enriched in organisms and their spatial distribution is discontinuous; therefore, nonproton nucleus images must be combined with h-MRI. 1 H-image registration is used to obtain anatomical location information of aprotic nuclides. Typically, aprotic nuclei are associated with... 1 1H MRI acquires images of two nuclides sequentially, doubling or lengthening the total scan time. It's difficult to keep the scanned area stationary for extended periods, introducing motion artifacts. This data acquisition method increases scan time and potentially complicates image registration. To address these issues, researchers have proposed dual-nucleus simultaneous excitation and acquisition imaging methods. However, in these methods, the pulsed gradient magnetic field always acts on all nuclides simultaneously. Since the gyromagnetic ratios of different nuclides vary significantly, the images obtained by simultaneous excitation and acquisition methods for different nuclides have different fields of view (FOV) and slice thicknesses, scaled by their respective gyromagnetic ratios. Furthermore, most dual-nucleus simultaneous imaging methods… 1 The excitation pulse and echo time of H are consistent with those of the nonproton nucleus. To improve the signal-to-noise ratio of the nonproton nucleus, a short echo time imaging method is used. 1 H-image contrast control is limited.

[0004] The slice thickness and in-plane resolution of nonproton nucleus images are related to 1 The large difference in H is unfavorable for non-proton nucleus imaging and... 1 Registration and comparison between H images can even lead to incorrect localization of feature regions in non-proton nucleus images. 1 Risks in H-images. How to balance nonproton nucleus resolution, signal-to-noise ratio, and... 1 The issue of H-contrast is one that needs to be fully considered in dual-core synchronous MRI. Summary of the Invention

[0005] To address the aforementioned problems in existing technologies, this invention provides a method for Cartesian sampling dual-core synchronous magnetic resonance imaging and image reconstruction, specifically for proton (… 1 H) In dual-nucleus simultaneous imaging with and without proton nuclei, the slice thickness and field of view (FOV) of the aproton nuclei can be adjusted according to the specific application scenario, thereby adjusting the resolution in the three spatial directions and obtaining different echo time-weighted resolutions. 1 H-image, improve 1 H image contrast control flexibility.

[0006] The above-mentioned objective of the present invention is achieved through the following technical solution:

[0007] A Cartesian sampling method for dual-core synchronous magnetic resonance imaging and image reconstruction includes the following steps:

[0008] Step 1: Selectively excite adjacent protons sequentially 1 H and non-proton nuclide X nuclei, using parameter ap1 to adjust the intensity of the layer-selective gradient pulse g4 of the X nuclei, to achieve X nuclei layer thickness Th X The adjustment, in 1 A pre-compensated gradient pulse g3 is applied along the plane direction between the excitation pulses of H and X nuclides, causing the refocusing plane to converge. 1 The dispersed phase of H;

[0009] Step 2: Time-sharing and sequential processing of X core and 1 Frequency encoding of the H magnetic resonance signal is performed, and the intensity of the non-proton nucleus frequency encoding gradient pulse is adjusted using parameter ap2 to achieve the field of view (FOVr) of the X-nucleus frequency encoding direction. X After adjustment, a frequency-coded compensation gradient pulse g9 is applied to make... 1 The k-trajectory in the H-frequency coding direction evolves to the target point, and then frequency-coded gradient pulses are applied twice in succession. 1 H is frequency-coded twice;

[0010] Step 3: Time-sharing and sequential processing of X core and 1 Phase encoding is performed on the magnetic resonance signal of H. The intensity of the X-core phase encoding gradient pulse is adjusted using parameter ap3 to achieve the field of view (FOVp) of the X-core phase encoding direction.X After adjustment, a phase-encoded compensation gradient pulse g13 is applied to make 1 The k-trajectory in the H-phase encoding direction evolves to the target point, and finally, the phase-encoded re-convergence gradient pulse g14 is applied for re-convergence. 1 The phase encoding directions of the two nuclides, H and X, are exhibited as a dispersion;

[0011] Step 4: Time-sharing sequential acquisition of X-core and 1 Magnetic resonance signal of H: While applying the X-nucleus frequency-coded gradient pulse g8, the echo signal echo1 of the X-nucleus with an echo time of TE1 is acquired; while applying 1 Simultaneously with the first frequency encoding gradient pulse g10 of H, the echo time is TE2. 1 The first echo signal of H, echo2; is applied... 1 Simultaneously with the second frequency encoded gradient pulse g11 of H, the echo time is acquired for TE2. 1 The second echo signal of H is echo3;

[0012] Step 5: Combine the X-core data collected in Step 4 with... 1 The magnetic resonance signal of H was filled into the X nucleus and 1 Image reconstruction is performed in the k-space of H.

[0013] Furthermore, the specific method for step one is as follows:

[0014] (1) Settings 1 The inspection field of view (FOV), layer thickness (Th), sampling spectral width (SW) of the receiving channel, and sampling matrix (MN) of H are defined, where M is the number of sampling points, N is the number of coding steps in the phase coding direction, and the inspection field of view (FOV) includes the frequency coding direction field of view (FOVr) and the phase coding direction field of view (FOVp).

[0015] (2) Set the parameters for channels f1 and f2, with channel f1 set to... 1 H corresponds to the RF excitation channel, and f2 is the RF excitation channel corresponding to the X core. The calculation is based on the parameters of channels f1 and f2, as well as parameter ap1. 1 The intensity of H-core selection gradient pulse g2 and the intensity of X-core selection gradient pulse g4;

[0016] (3) Set the parameters of the selected layer gradient channel Gs, specifically: the intensity and width of the layer damage gradient pulse g1. 1 Width and intensity of H-selective gradient pulse g2 (Gslevel1), width and intensity of pre-compensated gradient pulse g3, width and intensity of X-core selective gradient pulse g4 (Gslevel2), and width and intensity of selective re-aggregation gradient pulse g6.

[0017] The width of g2 and 1The RF excitation pulse width p1 of H is consistent with that of the X core, and the width of g4 is consistent with that of the RF excitation pulse width p2 of the X core.

[0018] The sum of the areas of g3, half the area of ​​g2, and half the area of ​​g4 is zero. The area of ​​a positive gradient is positive, and the area of ​​a negative gradient is negative. The area of ​​g6 is equal to half the area of ​​g4, and the polarity of the gradient of g6 is opposite to that of g4.

[0019] In terms of timing, g1, g2, g3, g4, and g6 are applied sequentially; and g2 is related to... 1 The center of the radio frequency excitation pulse of H is aligned, and the center of the radio frequency excitation pulse of g4 is aligned with that of the X core.

[0020] Furthermore, five gradient pulses, g1, g2, g3, g4, and g6, are applied in a bridged manner.

[0021] Furthermore, the calculation methods for the strengths of g2 and g4 are as follows:

[0022] The specific parameters for setting the f1 and f2 channels are as follows:

[0023] The parameters of the f1 channel include: 1 H's radio frequency excitation pulse, pulse width p1, and pulse power;

[0024] The parameters of channel f2 include: the radio frequency excitation pulse of the X core, the pulse width p2, and the pulse power;

[0025] The RF excitation pulse for channel f1 is applied before the RF excitation pulse for channel f2;

[0026] according to 1 The frequency domain width BW1 of the excitation pulse is calculated from the RF excitation pulse properties of H and the pulse width p1.

[0027] The frequency domain width BW2 of the excitation pulse is calculated based on the excitation radio frequency pulse properties and pulse width p2 of the X core.

[0028] according to 1 gyrometry of H Calculate the layer thickness Th and BW1. 1 The intensity of the layer selection gradient pulse g2 of H is Gslevel1;

[0029] Based on the gyrorhythm ratio of the X nucleus Using BW2 and Gslevel1, calculate the intensity of the layer selection gradient pulse g4 and Gslevel2 of the X kernel.

[0030]

[0031] Where ap1 is the layer thickness Th of the X core. X Adjustment parameters, ap1 > 0.

[0032] Furthermore, the specific method for step two is as follows:

[0033] Configure the parameters of the frequency encoding gradient channel Gr, including the width and intensity of the frequency encoding preparation gradient pulse g7, the intensity and width of the X-core frequency encoding gradient pulse g8, and the width and intensity of the frequency encoding direction compensation gradient pulse g9. 1 The width and intensity of the first frequency encoded gradient pulse g10 of H, 1 The width and intensity of the second frequency-coded gradient pulse g11 of H, and the width and intensity of the frequency-coded direction-damped gradient pulse g12;

[0034] The width of g8 is set to the sampling time acqutime. ;

[0035] The width of g10 is set to acqutime; the width of g11 is set to acqutime.

[0036] The widths of g7, g9, and g12 are the same as the width of g6. The area of ​​g7 is equal to half the area of ​​g8, and the polarity of g7 is opposite to that of g8. The sum of the area of ​​g9, half the area of ​​g10, and half the area of ​​g8 is zero, and the polarity of g9 is opposite to that of g8. The polarity of g11 is opposite to that of g10.

[0037] according to 1 gyrometry of H Using the field of view (FOVr), sampling spectral width (SW), and parameter ap2, calculate the intensities of g8, g10, and g11;

[0038] In terms of timing, g7, g8, g9, g10, g11 and g12 are applied sequentially.

[0039] Furthermore, the strength calculation methods for g8, g10, and g11 are as follows:

[0040] according to 1 gyrometry of H The field of view (FOVr) and sampling spectral width (SW) are used to calculate the... 1 H's frequency-coded gradient pulse intensity Grlevel1,

[0041] Calculate the frequency-coded gradient pulse intensity Grlevel2 of kernel X.

[0042]

[0043] Where ap2 is the field of view (FOVr) of the X-core frequency coding direction. X The adjustment parameter, FOVr X ap2 > 0;

[0044] The intensity of g8 is set to Grlevel2, the intensity of g10 is set to -Grlevel1, and the intensity of g11 is set to Grlevel1.

[0045] Furthermore, g11 and g12 are applied in a bridging manner.

[0046] Furthermore, the specific method for step three is as follows:

[0047] Set the parameters of the phase-encoded gradient channel Gp, including the width and intensity of the X-core phase-encoded gradient pulse g5, the intensity and width of the phase-encoded compensation gradient pulse g13, and the width and intensity of the phase-encoded retraction gradient pulse g14.

[0048] The widths of g5, g13, and g14 are consistent with the width of g6, and the sum of the areas of g5 and g13 satisfies the requirements of Shannon's sampling theorem. 1 The sum of the areas of the H-phase encoded gradient, g5, g13, and g14 is zero;

[0049] according to 1 gyrometry of H Given the phase encoding steps N, field of view (FOVp), and parameter ap3, calculate the intensities of g5, g13, and g14.

[0050] In terms of timing, g5, g13, and g14 are applied sequentially.

[0051] In terms of timing, g7, g5, and g6 are center-aligned, g13 and g9 are center-aligned, and g14 and g12 are center-aligned.

[0052] Furthermore, the strength calculation methods for g5, g13, and g14 are as follows:

[0053] according to 1 gyrometry of H The phase encoding steps N and the field of view (FOVp) are used to calculate... 1 The phase-encoded gradient intensity array Gplevel1(n) of H, n=1,2,…N;

[0054] Calculate the phase-encoded gradient intensity array Gplevel2(n) of kernel X, n=1,2,…N;

[0055]

[0056] Among them, ap3 is the field of view (FOVp) of the X-core phase encoding direction. X The adjustment parameter, FOVp X ap3 > 0;

[0057] The intensity of g5 is set to Gplevel2(n), n=1,2,…N; the intensity of g13 is set to -(Gplevel2(n)-Gplevel1(n)), n=1,2,…N; the intensity of g14 is set to Gplevel2(n), n=1,2,…N.

[0058] Furthermore, step four also includes setting the parameters of the receiving channel, including the sampling spectral width and the number of sampling points of the X-core receiving channel. The sampling spectral width of the X-core receiving channel is related to... 1 The sampling spectral width SW of the H receiving channel remains consistent, and the number of sampling points of the X core receiving channel is the same as... 1 The number of sampling points M in the H receiving channel remains constant.

[0059] Furthermore, an adjustable delay parameter d1 is set between g7 and g8 to adjust the echo time TE1 of the X core; an adjustable delay parameter d2 is set between g9 and g10 to adjust... 1 The echo time TE2 of H; an adjustable delay parameter d3 is set between g10 and g11 to adjust... 1 The echo time of H is TE3.

[0060] Furthermore, the specific method for step five is as follows:

[0061] The echo1 data of the X-core is filled into the k-space of the X-core for image reconstruction to obtain the image of the X-core; 1 The first echo signal (echo2) data of H is filled into the k-space corresponding to TE2, the data is flipped along the frequency encoding direction, and then image reconstruction is performed to obtain... 1 H's TE2-weighted image; 1 The second echo signal (echo3) of H is filled into the k-space corresponding to TE3 for image reconstruction, obtaining... 1 H is a TE3-weighted image.

[0062] Compared with the prior art, the present invention has the following advantages:

[0063] 1. The dual-nucleus synchronous imaging method proposed in this invention provides two independent perturbation gradient echo sequences for each nuclide. The flip angle and echo time can be controlled independently, which enriches the contrast adjustment. The flip angle is controlled by pulse power.

[0064] 2. The resolution of the X nuclide is adjustable, and the resolution can be optimized according to the specific application scenario. This alleviates the problem of large differences in the thickness and planar resolution of the two nuclide layers in the synchronous excitation and synchronous acquisition methods, thereby alleviating the problem of difficult image registration and the possibility of incorrect positioning of the X nuclide feature region.

[0065] 3. The imaging method of sequential excitation and time-division sampling reduces the requirements for the synchronous control of radio frequency pulses in different frequency bands in MRI spectrometers; time-division excitation and sampling do not require the radio frequency coils of the two nuclides to be in resonance at the same time, which can reduce the complexity of decoupling design between proton and X nuclide coil units, and make it easier to modify or design a system that can perform dual-nuclide synchronous imaging.

[0066] 4. For protons in the planar direction and frequency encoding direction, the first moment of the gradient is zero, which satisfies the condition for flow compensation. This can eliminate motion artifacts caused by cerebrospinal fluid and slow-flowing blood, thus possessing flow compensation function. Attached Figure Description

[0067] Figure 1 This is a timing diagram of a magnetic resonance imaging pulse sequence for one scan cycle (TR) according to an embodiment of the present invention, with the horizontal axis representing the time axis. Detailed Implementation

[0068] To facilitate understanding and implementation of the present invention by those skilled in the art, the present invention will be further described in detail below with reference to embodiments. It should be understood that the embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0069] Loaded into magnetic resonance imaging systems that support two or more radio frequency channels for transmission and reception, such as Figure 1 The pulse sequence shown; where two transmit channels and two receive channels are respectively connected to protons. 1 H corresponds to the non-proton X nucleus; the Sliceselection loop represents the layer selection loop; and the Phase-encoding loop represents the phase encoding loop. and They are respectively 1 The gyrometry ratio of H and X nuclei.

[0070] like Figure 1 As shown, a Cartesian sampling dual-nuclear synchronous magnetic resonance imaging and image reconstruction method includes the following steps:

[0071] Step 1: Select adjacent excitation protons. 1 H and non-proton nuclide X nuclei, using parameter ap1 to adjust the intensity of the layer-selective gradient pulse g4 of the X nuclei, achieve layer thickness Th of the X nuclei. X The adjustment, in 1 A pre-compensated gradient pulse g3 is applied along the plane direction between the excitation pulses of H and X nuclides, causing the refocusing plane to converge. 1 The dispersed phase of H;

[0072] The specific method for step one is as follows:

[0073] (1) Settings 1 The inspection field of view (FOV) of H, the layer thickness Th, the sampling spectral width SW of the receiving channel ADC2, and the sampling matrix [MN], where M is the number of sampling points, N is the number of coding steps in the phase coding direction, and the inspection field of view (FOV) includes the frequency coding direction field of view (FOVr) and the phase coding direction field of view (FOVp).

[0074] (2) Set the parameters and calculate the parameters for channels f1 and f2. Channel f1 is... 1 H corresponds to the radio frequency excitation channel, and channel f2 is the radio frequency excitation channel corresponding to the X core.

[0075] Configure the parameters for the f1 channel, including: 1 The radio frequency excitation pulse, pulse width p1, and pulse power of H are specified. In this embodiment, a 5-lobe SINC pulse is selected for channel f1.

[0076] The parameters of channel f2 are set, including: the radio frequency excitation pulse of the X core, the pulse width p2, and the pulse power. In this embodiment, the Gauss pulse is selected for channel f2.

[0077] The SINC RF pulse for channel f1 is applied before the Gauss RF pulse for channel f2.

[0078] The parameter calculation process is as follows:

[0079] according to 1 The frequency domain width BW1 of the excitation is calculated from the SINC RF excitation pulse properties of H and the pulse width p1.

[0080] The excitation frequency domain width BW2 is calculated based on the Gaussian excitation RF pulse properties and pulse width p2 of the X core.

[0081] according to 1 gyrometry of H Calculate the layer thickness Th and BW1. 1 The intensity of the layer selection gradient pulse g2 of H is Gslevel1.

[0082]

[0083] Based on the gyrorhythm ratio of the X nucleus Using BW2 and Gslevel1, the intensity of the layer selection gradient pulse g4 and Gslevel2 of the X kernel are calculated.

[0084]

[0085] Where ap1 is the layer thickness Th of the X core. X Adjustment parameters, ap1 > 0;

[0086] (3) Set the parameters of the selected gradient channel Gs

[0087] The parameters of the selected gradient channel Gs include the intensity and width of the layer damage gradient pulse g1. 1 The width and intensity of the H-layer selection gradient pulse g2, the width and intensity of the layer direction pre-compensation gradient pulse g3, the width and intensity of the X-core layer selection gradient pulse g4, and the width and intensity of the layer selection re-aggregation gradient pulse g6.

[0088] 1 The width of the H-kernel selection gradient pulse g2 is set to p1, and the intensity is set to Gslevel1; the width of the X-kernel selection gradient pulse g4 is set to p2, and the intensity is set to Gslevel2.

[0089] The area of ​​the pre-compensated gradient pulse g3 and 1 The sum of half the area of ​​the H-core selected layer gradient pulse g2 is equal to half the area of ​​the X-core selected layer gradient pulse g4, and the gradient polarity of g3 is the same as that of g2; the area of ​​the layer regrouping gradient pulse g6 is equal to half the area of ​​the X-core selected layer gradient pulse g4, and the gradient polarity of g6 is opposite to that of g4.

[0090] In terms of timing, g1, g2, g3, g4, and g6 are applied sequentially, and g2 is related to... 1 H's SINC RF excitation pulse is center-aligned, and g4's Gauss RF excitation pulse is center-aligned with X's.

[0091] In this embodiment, five gradient pulses, g1, g2, g3, g4, and g6, are applied after being bridged together.

[0092] Step 2: Time-sharing and sequential processing of X core and 1 The magnetic resonance signal of H is frequency encoded, and the intensity of the X-core frequency encoding gradient pulse g2 is adjusted using parameter ap2 to achieve the field of view (FOVr) of the X-core frequency encoding direction. X After adjustment, a frequency-coded compensation gradient pulse g9 is applied to make... 1 The k-trajectory in the H-frequency coding direction evolves to the target point, and then frequency-coded gradient pulses are applied twice in succession. 1 H is frequency-coded twice;

[0093] The specific method for step two is as follows:

[0094] (1) According to 1 gyrometry of H The field of view (FOVr) and sampling spectral width (SW) are used to calculate the... 1 H's frequency-coded gradient pulse intensity Grlevel1,

[0095]

[0096] Calculate the frequency-coded gradient pulse intensity Grlevel2 of kernel X.

[0097]

[0098] Where ap2 is the field of view (FOVr) of the X-core frequency coding direction. X The adjustment parameter, FOVr X ap2 > 0;

[0099] (2) Set the parameters of the frequency coding gradient channel Gr

[0100] The parameters of the frequency-coded gradient channel Gr include the width and intensity of the frequency-coded preparation gradient pulse g7, the intensity and width of the X-kernel frequency-coded gradient pulse g8, and the width and intensity of the frequency-coded direction compensation gradient pulse g9. 1 The width and intensity of the first frequency encoded gradient pulse g10 of H, 1 The width and intensity of the second frequency-coded gradient pulse g11 of H, and the width and intensity of the frequency-coded direction-damped gradient pulse g12.

[0101] The intensity of the X-core frequency-encoded gradient pulse g8 is set to Grlevel2, and the width is set to the sampling time acqutime, where acqutime ;

[0102] 1 The intensity of the first frequency encoded gradient pulse g10 of H is set to -Grlevel1, and the width is set to acqutime; 1 The intensity of the second frequency-coded gradient pulse g11 of H is set to Grlevel1, and the width is set to acqutime; the width of the frequency-coded direction-destructive gradient pulse g12 is consistent with the width of the layer-selective regrouping gradient pulse g6.

[0103] The width of the frequency encoding preparation gradient pulse g7 is consistent with the width of the layer selection regrouping gradient pulse g6, and the area is equal to half the area of ​​g8. The polarity of g7 is opposite to that of g8.

[0104] The width of the compensation gradient pulse g9 in the frequency coding direction is consistent with the width of the layer selection regrouping gradient pulse g6. The polarity of g9 is opposite to that of g8. The sum of the area of ​​g9 and half the area of ​​g10 is equal to half the area of ​​g8.

[0105] In terms of timing, g7, g8, g9, g10, g11, and g12 are applied sequentially. In this embodiment, g11 and g12 are applied after bridging.

[0106] Step 3: Time-sharing and sequential processing of X core and 1Phase encoding is performed on the magnetic resonance signal of H. The intensity of the X-core phase encoding gradient pulse g5 is adjusted using parameter ap3 to achieve the field of view (FOVp) of the X-core phase encoding direction. X After adjustment, a phase-encoded compensation gradient pulse g13 is applied to make 1 The k-trajectory in the H-phase encoding direction evolves to the target point, and finally, the phase-encoded re-convergence gradient pulse g14 is applied for re-convergence. 1 The phase encoding directions of the two nuclides, H and X, are exhibited as a dispersion;

[0107] The specific method for step three is as follows:

[0108] according to 1 gyrometry of H The phase encoding steps N and the field of view (FOVp) are used to calculate... 1 The phase-encoded gradient intensity array Gplevel1(n) of H, n=1,2,…N,

[0109] The phase-coded gradient pulse is transformed into an equivalent rectangular pulse, the duration of which is denoted as... ,according to 1 gyrometry of H Calculate the step size of the phase-coded gradient pulse intensity ,

[0110] Gplevel1(n) =

[0111] Calculate the phase-coded gradient pulse intensity array Gplevel2(n) of kernel X, n=1,2,…N;

[0112]

[0113] Among them, ap3 is the field of view (FOVp) of the X-core phase encoding direction. X The adjustment parameter, FOVp X ap3 > 0;

[0114] The parameters of the phase-encoded gradient channel Gp are set, including the width and intensity of the X-core phase gradient pulse g5, the intensity and width of the phase-encoded compensation gradient pulse g13, and the width and intensity of the phase-encoded regrouping gradient pulse g14.

[0115] The width of the X-core phase gradient pulse g5 is consistent with the width of the layer selection regrouping gradient pulse g6, and the intensity is set to Gplevel2(n), n=1,2,…N;

[0116] The width of the phase-encoded compensation gradient pulse g13 is consistent with the width of the layer-selective regrouping gradient pulse g6, and the intensity is set to -(Gplevel2(n)- Gplevel1(n)), n=1,2,…N;

[0117] The width of the phase-encoded reconvergence gradient pulse g14 is consistent with the width of the layer-selective reconvergence gradient pulse g6, and the intensity is set to Gplevel2(n), n=1,2,…N; an adjustable delay parameter d1 is set between g7 and g8 to adjust the echo time TE1 of the X-core; an adjustable delay parameter d2 is set between g9 and g10 to adjust… 1 The echo time TE2 of H; an adjustable delay parameter d3 is set between g10 and g11 to adjust... 1 H echo time TE3;

[0118] In terms of timing, g5, g13, and g14 are applied sequentially;

[0119] In terms of timing, g7, g5, and g6 are center-aligned, g13 and g9 are center-aligned, and g14 and g12 are center-aligned.

[0120] Step 4: Set the parameters of the receiving channel. The parameters of the receiving channel include the sampling spectral width SW and the number of sampling points M of the X-core receiving channel ADC1.

[0121] X-core and X-core were collected sequentially at different times. 1 Magnetic resonance signal of H: While applying the X-nucleus frequency-coded gradient pulse g8, the echo signal echo1 of the X-nucleus with an echo time of TE1 is acquired. 1 Simultaneously with the first frequency encoding gradient pulse g10 of H, the echo time is TE2. 1 The first echo signal of H, echo2, is applied... 1 Simultaneously with the second frequency encoded gradient pulse g11 of H, the echo time is acquired for TE2. 1 The second echo signal of H is echo3;

[0122] Step 5: Image Reconstruction

[0123] The echo signal (echo1) data of the X-core is filled into the k-space of the X-core for image construction, thus obtaining the image of the X-core; 1 The first echo signal (echo2) data of H is filled into the k-space corresponding to TE2, the data is flipped along the frequency encoding direction, and then image reconstruction is performed to obtain... 1 H's TE2-weighted image; 1 The second echo signal (echo3) of H is filled into the k-space corresponding to TE3 for image reconstruction, obtaining... 1 TE3 weighted image of H.

[0124] It should be noted that the specific embodiments described in this invention are merely illustrative of the spirit of the invention. Those skilled in the art to which this invention pertains can make various modifications or additions to the described specific embodiments or use similar methods to substitute them, without departing from the spirit of the invention or exceeding the scope defined by the appended claims.

Claims

1. A method for Cartesian sampling dual-core synchronous magnetic resonance imaging and image reconstruction, characterized in that, Includes the following steps: Step 1: Selectively excite adjacent protons sequentially 1 H and non-proton nuclide X nuclei, using parameter ap1 to adjust the intensity of the layer-selective gradient pulse g4 of the X nuclei, to achieve X nuclei layer thickness Th X The adjustment, in 1 A pre-compensated gradient pulse g3 in the planar direction is applied between the excitation pulses of the two nuclides H and X to regroup the dispersed phase of protons within the planar region. Step 2: Time-sharing and sequential processing of X core and 1 The magnetic resonance signal of H is frequency encoded, and the intensity of the X-core frequency encoding gradient pulse g8 is adjusted using parameter ap2 to achieve the field of view (FOVr) of the X-core frequency encoding direction. X After adjustment, a frequency-coded compensation gradient pulse g9 is applied to make... 1 The k-trajectory in the H-frequency coding direction evolves to the target point, and then frequency-coded gradient pulses are applied twice in succession. 1 H is frequency-coded twice; Step 3: Time-sharing and sequentially process the X core and 1 Phase encoding is performed on the magnetic resonance signal of H. The intensity of the X-core phase encoding gradient pulse g5 is adjusted using parameter ap3 to achieve the field of view (FOVp) of the X-core phase encoding direction. X After adjustment, a phase-encoded compensation gradient pulse g13 is applied to make 1 The k-trajectory in the H-phase encoding direction evolves to the target point, and finally, the phase-encoded re-convergence gradient pulse g14 is applied for re-convergence. 1 The phase encoding directions of the two nuclides, H and X, are exhibited as a dispersion; Step 4: Time-sharing sequential acquisition of X-core and 1 Magnetic resonance signal of H: While applying the X-nucleus frequency-coded gradient pulse g8, the echo signal echo1 of the X-nucleus with an echo time of TE1 is acquired; while applying 1 Simultaneously with the first frequency encoding gradient pulse g10 of H, the echo time is TE2. 1 The first echo signal of H, echo2; is applied... 1 Simultaneously with the second frequency encoded gradient pulse g11 of H, the echo time is acquired at TE3. 1 The second echo signal of H is echo3; Step 5: Combine the X-core data collected in Step 4 with... 1 The magnetic resonance signal of H was filled into the X nucleus and 1 Image reconstruction is performed in the k-space of H.

2. A method of dual-nuclear simultaneous magnetic resonance imaging and image reconstruction with Cartesian sampling according to claim 1, characterized in that, The specific method for step one is as follows: (1) Settings 1 The inspection field of view (FOV), layer thickness (Th), sampling spectral width (SW) of the receiving channel, and sampling matrix [MN] of H; where M is the number of sampling points, N is the number of coding steps in the phase coding direction, and the inspection field of view (FOV) includes the frequency coding direction field of view (FOVr) and the phase coding direction field of view (FOVp). (2) Set the parameters for channels f1 and f2, with channel f1 set to... 1 H corresponds to the RF excitation channel, and f2 is the RF excitation channel corresponding to the X core. The calculation is based on the parameters of channels f1 and f2, as well as parameter ap1. 1 The intensity of H-core selection gradient pulse g2 and the intensity of X-core selection gradient pulse g4; (3) Set the parameters of the selected layer gradient channel Gs, specifically: the intensity and width of the layer damage gradient pulse g1. 1 Width and intensity of H-selective gradient pulse g2, width and intensity of pre-compensated gradient pulse g3, width and intensity of X-core selective gradient pulse g4, and width and intensity of selective re-aggregation gradient pulse g6. The width of g2 is consistent with 1 The width of g4 is consistent with the radio frequency excitation pulse width p2 of the X nucleus. The sum of the areas of g3, half the area of ​​g2, and half the area of ​​g4 is zero. The area of ​​a positive gradient is positive, and the area of ​​a negative gradient is negative. The area of ​​g6 is equal to half the area of ​​g4, and the polarity of the gradient of g6 is opposite to that of g4. In terms of timing, g1, g2, g3, g4, and g6 are applied sequentially; and g2 is related to... 1 The center of the radio frequency excitation pulse of H is aligned, and the center of the radio frequency excitation pulse of g4 is aligned with that of the X core.

3. A dual-nuclear simultaneous magnetic resonance imaging and image reconstruction method with Cartesian sampling as claimed in claim 2, wherein, The calculation methods for the strengths of g2 and g4 are as follows: The specific parameters for the f1 and f2 channels are as follows: The parameters of the f1 channel include: 1 The radio frequency excitation pulse, pulse width p1, and pulse power of H; The parameters of channel f2 include: the radio frequency excitation pulse of the X core, the pulse width p2, and the pulse power; The RF excitation pulse for channel f1 is applied before the RF excitation pulse for channel f2; According to 1 The frequency domain width BW1 of the excitation pulse is calculated from the radio frequency excitation pulse properties and the pulse width p1 of H. The frequency domain width BW2 of the excitation pulse is calculated based on the excitation radio frequency pulse properties and pulse width p2 of the X core. according to 1 gyrometry of H Calculate the layer thickness Th and BW1. 1 The intensity of the selected gradient pulse g2 of H is Gslevel1; Based on the gyrorhythm ratio of the X nucleus Using BW2 and Gslevel1, calculate the intensity of the layer selection gradient pulse g4 Gslevel2 for the X kernel. wherein ap1 is the layer thickness Th of the X nucleus X adjustment parameters, ap1 >

0.

4. A dual-nuclear simultaneous magnetic resonance imaging and image reconstruction method of Cartesian sampling according to claim 3, characterized in that, The specific method for step two is as follows: Configure the parameters of the frequency encoding gradient channel Gr, including the width and intensity of the frequency encoding preparation gradient pulse g7, the intensity and width of the X-core frequency encoding gradient pulse g8, and the width and intensity of the frequency encoding direction compensation gradient pulse g9. 1 The width and intensity of the first frequency encoded gradient pulse g10 of H, 1 The width and intensity of the second frequency-coded gradient pulse g11 of H, and the width and intensity of the frequency-coded direction-damped gradient pulse g12; The width of g8 is set to the sampling time acqutime, acqutime ; The width of g10 is set to acqutime; the width of g11 is set to acqutime. The widths of g7, g9, and g12 are the same as the width of g6. The area of ​​g7 is equal to half the area of ​​g8, and the polarity of g7 is opposite to that of g8. The sum of the area of ​​g9, half the area of ​​g10, and half the area of ​​g8 is zero, and the polarity of g9 is opposite to that of g8. The polarity of g11 is opposite to that of g10. according to 1 gyrometry of H Using the field of view (FOVr), sampling spectral width (SW), and parameter ap2, calculate the intensities of g8, g10, and g11; In terms of timing, g7, g8, g9, g10, g11 and g12 are applied sequentially.

5. A method of dual-nuclear simultaneous magnetic resonance imaging and image reconstruction with Cartesian sampling according to claim 4, characterized in that, The strength calculation methods for g8, g10, and g11 are as follows: according to 1 gyrometry of H The field of view (FOVr) and sampling spectral width (SW) are used to calculate the... 1 H's frequency-coded gradient pulse intensity Grlevel1, Calculate the frequency-coded gradient pulse intensity Grlevel2 of kernel X. Where ap2 is the field of view (FOVr) of the X-core frequency coding direction. X The adjustment parameter, FOVr X ap2 > 0; The intensity of g8 is set to Grlevel2, the intensity of g10 is set to -Grlevel1, and the intensity of g11 is set to Grlevel1.

6. The method for dual-core synchronous magnetic resonance imaging and image reconstruction using Cartesian sampling according to claim 4, characterized in that, The specific method for step three is as follows: Set the parameters of the phase-encoded gradient channel Gp, including the width and intensity of the X-core phase-encoded gradient pulse g5, the intensity and width of the phase-encoded compensation gradient pulse g13, and the width and intensity of the phase-encoded retraction gradient pulse g14. The widths of g5, g13, and g14 are consistent with the width of g6, and the sum of the areas of g5 and g13 satisfies the requirements of Shannon's sampling theorem. 1 The sum of the areas of the H-phase encoded gradient, g5, g13, and g14 is zero; according to 1 gyrometry of H Given the phase encoding steps N, field of view (FOVp), and parameter ap3, calculate the intensities of g5, g13, and g14. In terms of timing, g5, g13, and g14 are applied sequentially. In terms of timing, g7, g5, and g6 are center-aligned, g13 and g9 are center-aligned, and g14 and g12 are center-aligned.

7. A dual-nuclear simultaneous magnetic resonance imaging and image reconstruction method of Cartesian sampling according to claim 6, characterized in that, The strength calculation methods for g5, g13, and g14 are as follows: according to 1 gyrometry of H The phase encoding steps N and the field of view (FOVp) are used to calculate... 1 The phase-encoded gradient intensity array Gplevel1(n) of H, n=1,2,…N; Calculate the phase-encoded gradient intensity array Gplevel2(n) of kernel X, n=1,2,…N; Among them, ap3 is the field of view (FOVp) of the X-core phase encoding direction. X The adjustment parameter, FOVp X ap3 > 0; The intensity of g5 is set to Gplevel2(n), n=1,2,…N; the intensity of g13 is set to -(Gplevel2(n)-Gplevel1(n)), n=1,2,…N; the intensity of g14 is set to Gplevel2(n), n=1,2,…N.

8. The method of claim 1, wherein the method is a dual-nuclear simultaneous magnetic resonance imaging and image reconstruction method with Cartesian sampling, characterized by, Step four also includes setting the parameters of the receiving channel, including the sampling spectral width and the number of sampling points of the X-core receiving channel. 1 The sampling spectral width SW of the H receiving channel remains consistent, and the number of sampling points of the X core receiving channel is the same as... 1 The number of sampling points M in the H receiving channel remains constant.

9. A dual-nuclear simultaneous magnetic resonance imaging and image reconstruction method of Cartesian sampling as claimed in claim 4, wherein, An adjustable delay parameter d1 is set between g7 and g8 to adjust the echo time TE1 of the X core; an adjustable delay parameter d2 is set between g9 and g10 to adjust... 1 The echo time TE2 of H; an adjustable delay parameter d3 is set between g10 and g11 to adjust... 1 The echo time of H is TE3.

10. A method for dual-core synchronous magnetic resonance imaging and image reconstruction using Cartesian sampling according to any one of claims 1-9, characterized in that, The specific method for step five is as follows: Fill the echo signal (echo1) data of the X kernel into the k-space of the X kernel for image reconstruction to obtain the image of the X kernel; 1 The first echo signal (echo2) data of H is filled into the k-space corresponding to TE2, the data is flipped along the frequency encoding direction, and then image reconstruction is performed to obtain... 1 H's TE2-weighted image; 1 The second echo signal (echo3) of H is filled into the k-space corresponding to TE3 for image reconstruction, obtaining... 1 H is a TE3-weighted image.

Images