Ultra-high resolution functional imaging method in an ultra-high field magnetic resonance system

Abstract

The application discloses a method for high-resolution functional imaging in a high-field magnetic resonance system. The method comprises the following steps: S1, determining a two-dimensional spatially selective radio frequency pulse sequence applied to a brain of a subject organism, the two-dimensional spatially selective radio frequency pulse sequence is used for reducing a field of view (FOV) in a phase encoding direction and not generating signal folding. Preferably, the method further comprises the following step: S2, performing under-sampling and reconstruction on K-space based on an EPI-based GRAPPA algorithm. The application has the beneficial effect of realizing high-resolution functional imaging in a high-field magnetic resonance system.

Description

Technical Field

[0001] This invention relates to an ultra-high resolution functional imaging method in an ultra-high field magnetic resonance system. Background Technology

[0002] Functional magnetic resonance imaging (fMRI) uses echo-planar imaging (EPI) sequences to image the brain of a subject. The image signal is affected by the blood oxygen level-dependent (BOLD) effect and varies over time. In practical applications, mathematical modeling and statistical analysis of the time-varying BOLD signal allows for quantitative research on the brain cortex and related neural activity mechanisms.

[0003] Ultra-high field magnetic resonance (UHF) systems, with their high signal-to-noise ratio, are widely used in animal imaging and play a crucial role in clinical medicine and biological functional research. However, UHF systems have yet to achieve ultra-high resolution imaging. The reasons are twofold: First, functional magnetic resonance imaging (fMRI) essentially involves statistical analysis of image signal changes over time, requiring repeated imaging to acquire a time series of images. These repetitions often reach hundreds or thousands, resulting in very long imaging times. Ultra-high resolution imaging demands large receiver bandwidth, high gradient strength, and numerous phase encoding operations. This leads to excessively high gradient waveform duty cycles within the repetition time (TR) period, resulting in excessive heat generation in the gradient system. Hardware limitations prevent long-duration fMRI. Second, due to the influence of UHF magnetic resonance properties on the BOLD signal, the effective echo time (TE) for functional imaging is relatively short. Besides requiring high receiver bandwidth and high readout gradient strength, the number of phase encoding operations must be limited to achieve a shorter TE. This contradicts the high number of imaging encoding operations required for ultra-high resolution imaging. Third, if we want to ensure ultra-high resolution without increasing the number of imaging encodings, then besides using parallel imaging techniques, the only option is to reduce the imaging field of view (FOV). However, when the FOV in the imaging encoding direction is smaller than the length of the excited subject, a folding phenomenon will occur because the receiver bandwidth is too small and the Nyquist sampling theorem is not satisfied. That is, signals outside the FOV will be folded into the image. Summary of the Invention

[0004] The technical problem to be solved by the present invention is that ultra-high field magnetic resonance systems cannot achieve ultra-high resolution functional imaging, and the present invention provides an ultra-high field magnetic resonance system ultra-high resolution functional imaging method.

[0005] To achieve this objective, the technical solution of the present invention is as follows: a method for ultra-high resolution functional imaging in an ultra-high field magnetic resonance system, comprising:

[0006] Step S1: Determine the two-dimensional spatially selective radio frequency pulse sequence to be applied to the brain of the subject organism. The two-dimensional spatially selective radio frequency pulse sequence is used to reduce the scan field FOV in the phase encoding direction and does not produce signal folding.

[0007] Step S2: The K-space is undersampled and reconstructed based on the GRAPPA (GeneRalizedAutocalibratingPartiallyParallelAcquisitions) algorithm of EPI (EchoPlanarImaging).

[0008] Compared with the prior art, the beneficial effects of the present invention are at least as follows: 1. The purpose of setting step S1 is to reduce the scanning field FOV in the phase encoding direction by designing a two-dimensional spatially selective radio frequency pulse sequence to selectively excite the imaging layer without causing image folding, thus making it possible to achieve ultra-high resolution imaging in an ultra-high field magnetic resonance system; 2. The purpose of setting step S2 is to reduce the number of phase encodings, further reduce the gradient duty cycle and effective echo time of functional imaging, so that the heat generation power of the gradient system is sufficiently small, thus making it possible to achieve ultra-high resolution imaging in an ultra-high field magnetic resonance system; 3. Setting steps S1 and S2 simultaneously makes the realization of ultra-high resolution imaging in an ultra-high field magnetic resonance system a reality.

[0009] In addition to the technical problems solved by the present invention, the technical features constituting the technical solutions, and the beneficial effects brought about by the technical features of these technical solutions as described above, other technical problems that the present invention can solve, other technical features contained in the technical solutions, and the beneficial effects brought about by these technical features will be further described in detail with reference to the accompanying drawings. Attached Figure Description

[0010] Figure 1 This is a flowchart of an embodiment of the present invention.

[0011] Figure 2 This is a schematic diagram of the design process of two-dimensional spatially selective radio frequency pulses in an embodiment of the present invention.

[0012] Figure 3 This is a schematic diagram of the transverse magnetization vector magnitude space excited by two-dimensional spatial selective radio frequency pulses in an embodiment of the present invention.

[0013] Figure 4 This is an example of ultra-high resolution functional imaging of rats in an embodiment of the present invention. Detailed Implementation

[0014] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings. It should be noted that these descriptions of embodiments are intended to aid in understanding the invention and do not constitute a limitation thereof. Furthermore, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0015] In this embodiment, the selected hardware environment is a Bruker 7.0T magnetic resonance imaging (MRI) device (model BioSpec70 / 20USR), and the software environment is ParaVision 6.0.1. It should be noted that the selected hardware environment is not limited to a 7.0T MRI device. In other embodiments, 9.4T or 11.7T MRI devices can be used.

[0016] In this embodiment, the selected test organism is a rat. It should be noted that the test organism is not limited to rats. In other embodiments, other rodents or other smaller animals may be used.

[0017] Please see Figure 1 The figure illustrates an ultra-high field magnetic resonance imaging (MRI) system with ultra-high resolution functional imaging. The imaging method consists of steps S1 and S2. Step S1 involves determining the two-dimensional spatially selective radio frequency (RF) pulse sequence to be applied to the brain of the subject. This RF pulse sequence is used to reduce the field of view (FOV) in the phase-encoding direction without causing signal folding. Step S2 involves undersampling and reconstruction of the K-space based on the GRAPPA (Gene Ralized Autocalibrating Partially Parallel Acquisitions) algorithm using EPI (Echo Planar Imaging).

[0018] The following is a detailed explanation of steps S1 and S2:

[0019] (1) Quantitative design of two-dimensional spatially selective radio frequency pulses.

[0020] By considering the rat brain size, slice resolution, and signal-to-noise ratio, a specific two-dimensional spatially selective radio frequency (RF) pulse was designed using the small-angle RF pulse K-space analysis method. Based on the theory proposed by John Pauly et al. in 1989, for the excitation waveform of a small-angle RF pulse, neglecting the effects of relaxation and partial resonance, its excitation waveform is approximately equal to the Fourier transform of the weighted function W(t) under the K-space trajectory K(t) of the RF pulse. The RF pulse K-space is defined as the integral of the applied gradient intensity G(t) over time when the RF pulse is emitted, as shown in the equation. The ratio of the weighting function W(t) to the absolute value of the gradient intensity |G(t)| is shown in the equation. The transverse magnetization vector Mxy(r) excited by the radio frequency pulse at space r is approximately the Fourier transform of W(t) in K space, see M xy (r)=γM0∫ K W(K)e ir*K dK. Where γ is a constant gyromagnetic ratio and M0 is a constant longitudinal magnetization vector at time zero. Using the above theory, if when transmitting an RF pulse, an alternating positive and negative gradient waveform is applied in the layer direction, where the gradient intensity is fixed at Gs (unit: Hz / mm) and the effective gradient time is T (unit: ms), and the phase encoding direction is a time gradient pulse within the alternating positive and negative gradient interval in the layer direction, where the gradient intensity is an array of Gp[i] and the effective gradient time is fixed at τ, the RF pulse generates a K-space trajectory in the phase encoding direction. The two-dimensional spatially selective RF pulse is composed of a series of sub-pulse excitations, the number of sub-pulse is consistent with the number of gradient waveforms in the layer direction, and the sub-pulse pulse width is fixed at d (unit: ms), and the sub-pulse intensity is an array of B1[i]. According to the small-angle RF excitation K-space analysis method, the selected layer thickness FOXs is determined by the layer direction gradient intensity Gs and the sub-pulse unit time excitation bandwidth TBWs (unit: Hz / ms), see equation FOX. s =TBW s / (d*G s The unit-time excitation bandwidth TBWp of the waveform constructed from the phase-encoded direction gradient intensity array Gp[i], the gradient effective time fixed at τ, and the sub-pulse intensity value array B1[i] determines the excitation range size FOXp of the phase-encoded direction, as shown in equation [equation]. Where nRF represents the number of sub-pulses.

[0021] Please see Figure 2 The figure shows a design schematic of this embodiment. Referring to equation FOX... s =TBW s / (d*G sThe sub-pulse excitation bandwidth per unit time TBWs = 4400 Hz / ms, the selected layer thickness FOXs = 1 mm, the effective gradient time T = 400 μs, and the layer direction gradient intensity Gs = 11 kHz / mm. The phase encoding direction excitation range size FOXp = 5 mm. The effective gradient time is fixed at τ = 200 μs, the gradient intensity array Gp[i] = [3.896, 3.896, 7.791, 3.896, 3.896, 9.090, 5.194, 5.194, 5.194, 5.194, 9.090, 3.896, 3.896, 7.791, 3.896, 3.896] kHz, the sub-pulse intensity value array B1[i] = [0.64, 0.99, 0.78, 1.00, 1.65, 1.40, 3.21, 6.42, 7.76, 6.42, 3.21, 1.40, 1.65, 1.00, 0.78, 0.99, 0.64] μT, and the FOV size is 24 × 6 mm. 2 The image matrix size is 120×30.

[0022] Please see Figure 3 The figure shows a schematic diagram of the transverse magnetization vector of two-dimensional selective excitation obtained by calculating the two-dimensional spatially selective radio frequency pulse using the above design.

[0023] (2) Use EPI’s GRAPPA algorithm to reduce the number of imaging codes.

[0024] The two-dimensional spatially selective radio frequency pulses designed in step (1) were applied to ultra-high resolution functional imaging of the rat brain. To further meet the requirements of effective echo time and gradient waveform duty cycle for ultra-high field magnetic resonance imaging, the GRAPPA parallel imaging technique of EPI was used to reduce the number of imaging codes. Unlike the traditional method of selecting the coding line near the center of K space as the ACS (AutoCalibrating Scan), the ACS in this embodiment uses the data acquired by multi-excitation EPI as the pre-scan method. Among them, the number of multiple excitations of multi-excitation EPI Nshot = 2 is set to be equal to the acceleration factor R of GRAPPA parallel imaging, so that the T2* attenuation condition of each echo coding in EPI is close, and the reconstruction accuracy of GRAPPA algorithm is higher.

[0025] (3) Phase correction of Ghost artifacts.

[0026] Ghost correction pre-scan for correcting the phase of odd and even echoes in EPI images involves setting the phase encoding gradient to zero during EPI imaging, subtracting the phase of the acquired odd and even echo data, and calculating the phase correction value using polynomial fitting.

[0027] (4) Phase correction for ACS pre-scan.

[0028] The phase correction value obtained from the Ghost correction pre-scan data in step (3) is used to perform phase correction on the ACS pre-scan data in step (2), and then the GRAPPA algorithm is used to fit the linear weighting factor of each block of each receiving coil channel.

[0029] (5) Reconstruction was performed using the GRAPPA algorithm.

[0030] The BOLD functional imaging data acquired in steps (1) and (2) are used to perform Ghost phase correction on the functional imaging data by calculating the correction value obtained from the pre-scan data in step (3). Then, the functional imaging data are reconstructed in parallel using GRAPPA by using the linear weighting factor calculated from the pre-scan data in step (4) to obtain an ultra-high resolution functional image.

[0031] Please see Figure 4 The figure shows ultra-high resolution functional images of rats acquired using the imaging method described in this embodiment on a 7.0T magnetic resonance imaging device, with the pixel size controlled below 200μm. It should be noted that in existing technologies, other imaging methods yield rat functional images with pixel sizes ranging from 300-500μm.

[0032] The above description merely illustrates embodiments of the present invention and is quite specific and detailed; however, it should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A method for ultra-high resolution functional imaging in an ultra-high field magnetic resonance system, characterized in that, The imaging method includes: step S1, determining a two-dimensional spatially selective radio frequency pulse sequence to be applied to the brain of the subject, wherein the two-dimensional spatially selective radio frequency pulse sequence is used to reduce the scanning field FOV in the phase encoding direction and does not produce signal folding; and step S2, performing undersampling and reconstruction of the K space based on the GRAPPA algorithm of EPI. Step S1 includes: the subject organism is a rat; using the small-angle radio frequency pulse K-space analysis method, a specific two-dimensional spatially selective radio frequency pulse is designed; the two-dimensional spatially selective radio frequency pulse sequence is configured as follows: sub-pulse unit time excitation bandwidth TBWs = 4400Hz / ms, layer thickness FOXs = 1mm, gradient effective time T = 400μs, layer direction gradient intensity Gs = 11KHz / mm; phase encoding direction excitation range FOXp = 5mm; gradient effective time fixed at τ = 200μs, gradient intensity array Gp[i] = [3.896, 3.896, 7.79...]. 1, 3.896, 3.896, 9.090, 5.194, 5.194, 5.194, 5.194, 9.090, 3.896, 3.896, 7.791, 3.896, 3.896]kHz, sub-pulse intensity value array B1[i]=[0.64, 0.99, 0.78, 1.00, 1.65, 1.40, 3.21, 6.42, 7.76, 6.42, 3.21, 1.40, 1.65, 1.00, 0.78, 0.99, 0.64]μT, FOV size is 24×6mm2, image matrix size is 120×30; Step S2 includes: sub-step S21, ACS pre-scan: using data acquired by multi-excitation EPI as the pre-scan method, with the number of multiple excitations Nshot of multi-excitation EPI set to be equal to the acceleration factor R of the GRAPPA method; sub-step S22, phase correction of Ghost artifacts; sub-step S23, phase correction of ACS pre-scan; and sub-step S24, parallel reconstruction of functional imaging data using the GRAPPA algorithm to obtain ultra-high resolution functional imaging.

Images