Single-coil multilayer parallel magnetic resonance imaging method based on multiple excitation in low field

Abstract

The invention discloses a single-coil multi-layer parallel magnetic resonance imaging method based on multiple excitation in a low field, and the method comprises the steps: achieving the multi-layer parallel excitation through a multi-band pulse, and carrying out the 0-pi phase modulation of the multi-band pulse; constructing a fast spin echo sequence of secondary phase modulation; performing imaging by using the fast spin echo sequence subjected to secondary phase modulation; and reconstruction is carried out through an odd-even echo classification method, and a scanning result graph is obtained. Aiming at the common problem of low signal to noise ratio of low-field magnetic resonance imaging in practical application, based on hardware characteristics of a single coil of low-field magnetic resonance equipment and actual requirements of repeated scanning, a series of SMS related technologies such as an FSE fast spin echo sequence, Hadamard pulse coding, multi-band pulse and secondary phase modulation are combined; and the signal-to-noise ratio of the low-field FSE sequence is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of magnetic resonance imaging, and in particular to a single-coil multi-layer parallel magnetic resonance imaging method based on multiple excitations under low field. Background Art

[0002] Magnetic resonance imaging (MRI) is a non-contact, non-invasive, and radiation-free medical imaging technology that can provide high-resolution soft tissue images and, through sequence design and parameter selection, can deliver multi-contrast results, offering more accurate diagnostic evidence for clinical diseases. With the development of MRI technology, low-field MRI equipment and related technologies, which offer lower field strengths and smaller sizes, have gradually attracted the attention of researchers. Low-field MRI equipment is inexpensive, easy to use, and has a wide range of applications. It also possesses favorable physical properties such as a lower specific absorption rate (SAR) and a longer T2 time, and thus holds enormous potential for development. However, since MRI signal intensity is proportional to the square of the magnetic field strength, a decrease in magnetic field strength will result in a decrease in MRI signal intensity. Therefore, insufficient signal-to-noise ratio is a common and significant issue facing low-field MRI technology and a key area of research currently underway for low-field MRI researchers.

[0003] Simultaneous multi-slice (SMS) is a parallel imaging technique that acquires multi-slice image signals in a single excitation. It uses multi-band pulses to simultaneously excite and acquire multi-slice signals, which are then decoded during reconstruction to produce the multi-slice image. In its early development, SMS significantly reduced imaging efficiency due to the increased number of scans required for encoding and decoding, leading to its low focus. However, with the advancement of multi-coil and parallel imaging technologies, SMS has gradually entered clinical application through integration with multi-coil parallel imaging and has been widely used in the field of accelerated acquisition. For example, SMS has been combined with phase-offset multiplanar volume imaging (POMP) to obtain two separate slices in a single scan. Compared to high-field MRI, low-field MRI equipment generally lacks the hardware requirements for multi-coil operation. However, the reduced signal-to-noise ratio associated with low field strength has led to the widespread adoption of repeated scanning strategies in low-field MRI. These limitations have limited mainstream SMS methods based on multi-coil and parallel imaging, bringing traditional SMS methods based on Hadamard encoding and multiple excitations back into the spotlight. Considering the efficiency improvement brought by SMS technology at the same number of scans and the maintenance of image signal-to-noise ratio, the application of this method at low field will help improve the signal-to-noise ratio efficiency of imaging and provide a new approach to solving the low-field signal-to-noise ratio problem.

[0004] The problems existing in the prior art are:

[0005] 1. The signal-to-noise ratio efficiency in low-field fast spin echo (FSE) sequences is generally low;

[0006] 2. The peak value of multi-band pulse B1 is too high under conventional design;

[0007] 3. Non-Carr-Purcell-Meiboom-Gill (CPMG) sequence conditions: Residual magnetization in the system leads to rapid attenuation of the magnetic resonance signal and image artifacts;

[0008] 4. Odd-even echo signal oscillation caused by secondary phase modulation.

[0009] Therefore, those skilled in the art are committed to developing a single-coil multi-slice parallel magnetic resonance imaging method based on multiple excitations at a low field to overcome the problems existing in the prior art. Summary of the Invention

[0010] In view of the above-mentioned defects of the prior art, the technical problem to be solved by the present invention is how to realize multi-layer parallel imaging based on the actual situation of low-field magnetic resonance imaging, and ultimately improve the signal-to-noise ratio of FSE sequence images.

[0011] To achieve the above object, the present invention provides a single-coil multi-slice parallel magnetic resonance imaging method based on multiple excitations at low field, comprising the following steps:

[0012] Step 1: using a multi-band pulse to achieve multi-layer parallel excitation, wherein the multi-band pulse is subjected to 0-π phase modulation;

[0013] Step 2: construct a quadratic phase modulated fast spin echo sequence;

[0014] Step 3, performing imaging using the quadratic phase modulated fast spin echo sequence;

[0015] Step 4: Reconstruct the image by classifying the odd and even echoes to obtain the scan result image.

[0016] Furthermore, the multi-band in step 1 is N-band, the phases of the N excitation pulses of the N-band correspond to the N-order Hadamard matrix, and N layers of images are acquired by N repeated excitations, where N is 4 or 8.

[0017] Furthermore, the N frequency bands are 4 frequency bands, and the phase of the corresponding refocusing pulse is [0, π, π, 0].

[0018] Furthermore, the N frequency bands are 8 frequency bands, and the phases of the corresponding refocusing pulses are [0,π,0,π,π,0,π,0].

[0019] Furthermore, the step 2 specifically includes the following steps:

[0020] Step 2.1, applying an excitation pulse and a corresponding refocusing pulse within a repetition time;

[0021] Step 2.2: Apply a layer selection gradient while applying each pulse;

[0022] Step 2.3: Apply two phase encoding gradients of equal magnitude and opposite direction between the two refocusing pulses.

[0023] Step 2.4: Apply a frequency encoding gradient between two phase encoding gradients while simultaneously acquiring the signal.

[0024] The emission phase of each refocusing pulse and the receiving phase of the corresponding signal satisfy the quadratic phase modulation method; and within one repetition time, multiple groups of signal excitation and acquisition with different level combinations are performed.

[0025] Furthermore, the step 2 further includes the following steps:

[0026] Add a destruction gradient before and after each layer selection gradient;

[0027] At the end of each repetition time, a return pulse is added.

[0028] Furthermore, the echo train length of the quadratic phase modulated fast spin echo sequence in step 2 is not less than 8.

[0029] Furthermore, the transmit phase of each refocusing pulse of the quadratic phase modulated fast spin echo sequence in step 2 satisfies the quadratic phase modulation method. Specifically, the receive phase Φ is defined as rx0 is 0 degrees, the emission phase of the nth refocusing pulse Φ txn Satisfy Φ txn =Φ rx(n-1) +δ n , the receiving phase Φ when receiving the signal for the nth time rxn Satisfy Φ rxn =Φ rxn +2·δ n , where δ satisfies δ n =δ n-1 +Δ n , Δ nWhen n is less than or equal to 7, it is a specific value, that is, Δ1 = 68.92°, Δ2 = 69.35°, Δ3 = 81.22°, Δ4 = 71.15°, Δ5 = 46.67°, Δ6 = 71.16°, Δ7 = 101.55°, Δ n When n is greater than or equal to 8, the value is fixed at 68.92°.

[0030] Furthermore, the step 3 specifically includes the steps of:

[0031] For the N-layer parallel excitation imaging process, the scan uses the corresponding N excitation pulses and the corresponding refocusing pulses to perform 2N imaging times, that is, the same phase encoding gradient is used in two adjacent repetition times, but the phase encoding values between adjacent odd and even echoes are exchanged, and this process is repeated N times to obtain the complete k-space of odd and even echo images respectively.

[0032] Furthermore, the step 4 specifically includes the steps of:

[0033] Step 4.1: For each of the two excitations of an excitation pulse, combine the data of all odd echoes into odd-echo k-space, and combine the data of all even echoes into even-echo k-space;

[0034] Step 4.2: Decode the combined odd echo data set and even echo data set according to the Hadamard coding principle and reconstruct them according to the traditional image reconstruction process. The two sets of images are combined to obtain a scanning result image.

[0035] The beneficial effects of the present invention are:

[0036] 1. Through SMS technology, multi-layer images can be obtained under the same scanning time, effectively improving the signal-to-noise ratio of FSE sequence images;

[0037] 2. Reduce the B1 peak of the pulse (especially the inversion pulse) through 0-π phase modulation and Shinnar-LeRoux Pulse (SLR) design method;

[0038] 3. By performing secondary phase modulation on the transmit phase of each refocusing pulse, an attenuation signal that stably oscillates between odd and even echoes is generated, significantly reducing artifacts caused by residual non-CPMG magnetization and improving image quality.

[0039] 4. Through two odd-even echo staggered scans, the complete k-space of odd and even echo images is collected, and the odd and even echo images are restored separately by classifying the echo signals to obtain a reconstructed image without odd and even echo artifacts.

[0040] The concept, specific structure and technical effects of the present invention will be further described below in conjunction with the accompanying drawings to fully understand the purpose, characteristics and effects of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a schematic diagram of the waveform and simulation of the first excitation pulse in a multi-band excitation pulse group according to a preferred embodiment of the present invention;

[0042] Figure 2 1 is a second excitation pulse waveform and simulation diagram in a multi-band excitation pulse group according to a preferred embodiment of the present invention;

[0043] Figure 3 1 is a schematic diagram of the waveform and simulation of the third excitation pulse in the multi-band excitation pulse group of a preferred embodiment of the present invention;

[0044] Figure 4 1 is a schematic diagram of the waveform and simulation of the fourth excitation pulse in the multi-band excitation pulse group according to a preferred embodiment of the present invention;

[0045] Figure 5 1 is a multi-band refocusing pulse waveform and simulation diagram of a preferred embodiment of the present invention;

[0046] Figure 6 Schematic diagram of the FSE pulse sequence used for imaging in a preferred embodiment of the present invention;

[0047] Figure 7 Schematic diagram of a method for setting the receiving phase and transmitting phase of a quadratic phase increment method according to a preferred embodiment of the present invention;

[0048] Figure 8 1 is a schematic diagram of odd-even echo correction according to a preferred embodiment of the present invention;

[0049] Figure 9 This is a result diagram before odd-even echo correction according to a preferred embodiment of the present invention;

[0050] Figure 10 This is a diagram showing the result of odd-even echo correction according to a preferred embodiment of the present invention;

[0051] Figure 11 This is a comparison diagram of imaging results using a multi-layer parallel excitation technology and a traditional single-layer excitation technology according to a preferred embodiment of the present invention;

[0052] Figure 12 This is a comparison diagram of the signal-to-noise ratio of imaging results using a multi-layer parallel excitation technology and a traditional single-layer excitation technology in a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0053] The following describes preferred embodiments of the present invention with reference to the accompanying drawings to make its technical content clearer and easier to understand. The present invention can be embodied in many different forms of embodiments, and the scope of protection of the present invention is not limited to the embodiments mentioned herein.

[0054] To address the generally low signal-to-noise ratio (SNR) efficiency in low-field fast spin echo (FSE) imaging sequences, this invention utilizes SMS technology adapted to low-field hardware conditions. This technology includes multi-band pulse design, sequence secondary phase modulation, and SMS image reconstruction. SMS technology enables the acquisition of multi-slice images within the same scan time, effectively improving the SNR of FSE sequence images within the same scan time.

[0055] In order to solve the problem of excessively high B1 peak of multi-band pulses under conventional design, the present invention reduces the B1 peak of pulses (especially flip pulses) through 0-π phase modulation and SLR pulse design method. The SLR pulse design method is used to obtain accurate layer selection profiles, and 0-π phase modulation is performed on different sub-bands to meet the conditions of the Karl-Purcell-Meibom-Gill (CPMG) sequence and reduce the B1 peak to obtain wide layer thickness and narrow interval layer selection pulses, and the B1 peak is controlled within Level

[0056] To address the problem of rapid attenuation of magnetic resonance signals and image artifacts caused by residual non-CPMG magnetization in the system, the present invention generates a stably attenuated echo signal through secondary phase modulation and odd-even echo classification. By performing secondary phase modulation on the transmit phase of each refocusing pulse, an attenuation signal that stably oscillates between odd and even echoes is generated, eliminating the influence of residual non-CPMG magnetization, thereby significantly reducing the artifact problem caused by residual non-CPMG magnetization and improving image quality.

[0057] To address the problem of odd-even echo signal oscillation caused by quadratic phase modulation, the present invention adopts an odd-even echo staggered scanning and classification reconstruction method. Through two odd-even echo staggered scans, the complete k-space of odd and even echo images is collected, and the odd and even echo images are restored separately by classifying the echo signals, so that a reconstructed image without odd-even echo artifacts can be obtained.

[0058] The purpose of the present invention is to realize multi-layer parallel imaging based on the actual situation of low-field magnetic resonance imaging, and ultimately improve the signal-to-noise ratio of FSE sequence images.

[0059] This technology is based on FSE sequence technology and achieves multi-layer parallel excitation through multi-band pulse design. The multi-band pulses used are phase modulated from 0 to π to meet CPMG conditions and reduce the B1 peak. Secondary phase modulation is performed on the FSE sequence to eliminate the influence of non-CPMG magnetization residue. The odd-even echo classification reconstruction method is used to solve the odd-even echo oscillation problem caused by secondary phase modulation. The specific steps are as follows:

[0060] (1) Design multi-band pulses;

[0061] (2) Design a quadratic phase modulated FSE sequence;

[0062] (3) Imaging using FSE sequence;

[0063] (4) Reconstruction through odd-even classification method.

[0064] Furthermore, the multi-band pulse in (1) is generally 4 bands or 8 bands. For the excitation pulse, the phase design of the N-band excitation pulse should correspond to the N-order Hadamard matrix, and N layers of images are acquired through N repeated excitations. For the refocusing pulse, the phase of the 4-band refocusing pulse should be [0,π,π,0], and the phase of the 8-band refocusing pulse should be [0,π,0,π,π,0,π,0].

[0065] Furthermore, the FSE sequence in (2) should be a quadratic phase modulated FSE sequence with an echo train length of no less than 8, and different echo times (EchoTime, TE) and repetition times (RepetitionTime, TR) are determined according to the actual contrast requirements. The transmit phase of each refocusing pulse of the quadratic phase modulated fast spin echo sequence satisfies the quadratic phase modulation method. Specifically, the receive phase Φ is defined as rx0 is 0 degrees, the emission phase of the nth refocusing pulse Φ txn Satisfy Φ txn =Φ rx(n-1) +δ n , the receiving phase Φ when receiving the signal for the nth time rxn Satisfy Φ rxn =Φ rxn +2·δ n , where δ satisfies δ n =δ n-1 +Δ n , Δ n When n is less than or equal to 7, it is a specific value, that is, Δ1 = 68.92°, Δ2 = 69.35°, Δ3 = 81.22°, Δ4 = 71.15°, Δ5 = 46.67°, Δ6 = 71.16°, Δ7 = 101.55°, Δ n When n is greater than or equal to 8, the value is fixed at 68.92°.

[0066] Furthermore, the imaging process in (3) should be performed using the FSE sequence in (2). For the N-layer parallel excitation imaging process, the scan uses the corresponding N excitation pulses and the corresponding refocusing pulses to perform 2N imaging times, that is, the same phase encoding gradient is used in two adjacent TRs, but the phase encoding values between adjacent odd and even echoes are exchanged, and this process is repeated N times to obtain the complete k-space of the odd and even echo images respectively.

[0067] Furthermore, the reconstruction in (4) should use all the data obtained by the scanning method in (3). During the reconstruction process, the data should be classified and reconstructed according to the odd and even positions of the echoes. Specifically, for the two excitations of each excitation pulse, the data of all odd echoes should be combined into the odd echo k-space, and the data of all even echoes should be combined into the even echo k-space; for the combined odd echo data group and the even echo data group, decoding is performed according to the Hadamard coding principle, and reconstruction is performed according to the traditional image reconstruction process. The two groups of images are combined to obtain the scanning result image.

[0068] Figure 1-5 The 4-band pulse waveform and selected layer profile designed for this invention are shown. The simulation uses 200 points arranged along the Z axis to simulate nuclei, with the initial magnetization intensity vector of each nucleus oriented in the positive direction of the Z axis.

[0069] Figure 1 It is the first excitation pulse of the excitation pulse group, and its corresponding code is

[1111] . After applying this pulse, the magnetization vector of the nuclei in the selected layer range in the Y direction becomes 1, and the magnetization vector in the Z direction becomes 0. Figure 1 In the figure, the upper left figure is the amplitude waveform of the excitation pulse, the upper right figure is the simulation result of the layer selection profile in the X-axis direction, the lower left figure is the simulation result of the layer selection profile in the Y-axis direction, and the lower right figure is the simulation result of the layer selection profile in the Z-axis direction.

[0070] Figure 2 This is the second excitation pulse in the excitation pulse group, and its corresponding code is [1-11-1]. After applying this pulse, the magnetization vector of the nuclei in the selected layer range in the Y direction becomes 1, and the magnetization vector in the Z direction becomes 0. Figure 2 In the figure, the upper left figure is the amplitude waveform of the excitation pulse, the upper right figure is the simulation result of the layer selection profile in the X-axis direction, the lower left figure is the simulation result of the layer selection profile in the Y-axis direction, and the lower right figure is the simulation result of the layer selection profile in the Z-axis direction.

[0071] Figure 3 This is the third excitation pulse in the excitation pulse group, and its corresponding code is [11-1-1]. After applying this pulse, the magnetization vector of the nuclei in the selected layer range in the Y direction becomes 1, and the magnetization vector in the Z direction becomes 0. Figure 3In the figure, the upper left figure is the amplitude waveform of the excitation pulse, the upper right figure is the simulation result of the layer selection profile in the X-axis direction, the lower left figure is the simulation result of the layer selection profile in the Y-axis direction, and the lower right figure is the simulation result of the layer selection profile in the Z-axis direction.

[0072] Figure 4 This is the fourth excitation pulse in the excitation pulse group, and its corresponding code is [1-1-1-1]. After applying this pulse, the magnetization vector of the nuclei in the selected layer range in the Y direction becomes 1, and the magnetization vector in the Z direction becomes 0. Figure 4 In the figure, the upper left figure is the amplitude waveform of the excitation pulse, the upper right figure is the simulation result of the layer selection profile in the X-axis direction, the lower left figure is the simulation result of the layer selection profile in the Y-axis direction, and the lower right figure is the simulation result of the layer selection profile in the Z-axis direction.

[0073] Figure 5 In order to refocus the pulse, when the pulse is generated, the phase values of the four frequency bands are [0ππ0] respectively. After the pulse is applied, the magnetization vector of the nuclei in the selected layer range in the Z direction becomes -1. Figure 5 In the figure, the upper right figure is the simulation result of the layer selection profile in the X-axis direction, the lower left figure is the simulation result of the layer selection profile in the Y-axis direction, and the lower right figure is the simulation result of the layer selection profile in the Z-axis direction.

[0074] Figure 6 The FSE scanning sequence designed and adjusted for the present invention. The sequence diagram shows the sequence timing within a TR, which corresponds to the excitation pulse emission timing, the layer selection gradient application timing, the phase encoding application timing, the frequency encoding application timing and the echo signal timing from top to bottom. Within a TR time, an excitation pulse is applied, followed by N excitation pulses; while each pulse is applied, a layer selection gradient is applied; between the two excitation pulses, two phase encoding gradients of equal size and opposite direction are applied; between the two phase encoding gradients, a frequency encoding gradient is applied, and the signal is collected at the same time. Before and after each layer selection gradient, a crusher gradient is added. At the end of each TR, a return pulse (Tipup) module is added. The emission phase of each refocusing pulse and the receiving phase of the corresponding signal meet the quadratic phase modulation method.

[0075] Within a single repetition time, multiple slice combinations can be excited and acquired. For example, if there are 16 slices,

[15913] represents the first slice combination, and [261014] represents the second slice combination, then, if the repetition time is long enough, the second slice combination can be acquired after the first one. Each slice combination is excited and acquired only once within a single repetition time.

[0076] Figure 7is a secondary phase modulation method for transmission and reception. The angle between the dotted line and the horizontal line in the figure corresponds to the transmission phase of the refocusing pulse, and the angle between the solid line and the horizontal line corresponds to the reception phase. The transmission phase of the nth refocusing pulse is Φ txn , the receiving phase when receiving the signal for the nth time is Φ rxn , define Φ rx0 is 0 degrees. Then Φ txn Satisfy Φ txn =Φ rx(n-1) +δ n The calculation method of Φ rxn Satisfy Φ rxn =Φ rxn +2·δ n The calculation method of δ satisfies δ n =δ n-1 +Δ n The calculation method of Δ is shown in Table 1.

[0077]

[0078] Table 1

[0079] Figure 8 The left figure shows the actual k-space sampling order, with the same row representing one echo and the same column representing one scan. The right figure shows the rearranged k-space diagram. Lines of the same color in the two images represent the same data. The odd-even echo correction method used in the present invention swaps the phase encoding corresponding to the odd and even echoes in two consecutive scans of the actual scan, and the arrangement order is shown in the left figure. During the reconstruction process, all odd echoes and all even echoes in the two scans are classified into two k-spaces respectively to form a completely sampled odd-even echo k-space to form an image.

[0080] Figure 9 The four images in the same row correspond to the four simultaneously excited layers of the same scan. The phase encoding of scan 1 and scan 2 corresponds to Figure 8 The first and second samples in .

[0081] Figure 10 For the corrected results, the first row is the reconstructed image generated using all odd echoes, and the second row is the reconstructed image generated using all even echoes.

[0082] Based on the 110mT self-developed low-field portable experimental platform, an imaging experiment was conducted on a healthy subject according to the principles and methods of the present invention. The imaging method used was the FSE sequence adjusted and designed by the present invention, and two experiments were conducted using single-band pulses and multi-band pulses respectively. The imaging field of view was 240mm×240mm, the excitation layer thickness was 5mm, the TR was 600ms, and the TE was 14.5ms (where the TR time length is the interval time between repeated excitations on the same layer, i.e., the repetition time; the TE length is the time from the 90° pulse to the echo center). The echo chain length was 16, that is, 16 180° pulses were applied after each 90° pulse, and 16 echo signals were collected accordingly. The sampling matrix size was 128×128, and the number of layers was 8.

[0083] Figure 11 The top row shows the results of single-band imaging compared to the SMS imaging results of the present invention. The bottom row shows the results of imaging using the traditional single-layer technique, and the top row shows the results of imaging using the multi-layer parallel excitation technique of the present invention. Each column corresponds to a layer in the imaging.

[0084] Through the technology of the present invention, under the condition of 8 bands, the imaging signal-to-noise ratio can be obtained to be about The improvement of the median value is reflected in the experimental results as a 9dB increase in the signal-to-noise ratio. Figure 12 Shown Figure 11 Signal-to-noise ratio image of the mid-scan results. Figure 12 The grayscale values in the results correspond to the signal-to-noise ratio of the region. The closer to white, the higher the signal-to-noise ratio. The top row shows the signal-to-noise ratio images of a traditional single-layer technique, while the bottom row shows the signal-to-noise ratio images of the multi-layer parallel excitation technique of the present invention. Each column corresponds to a layer in the image.

[0085] The present invention addresses the common problem of low signal-to-noise ratio in practical applications of low-field magnetic resonance imaging. Based on the hardware characteristics of the single coil of low-field magnetic resonance equipment and the actual needs of repeated scanning, it combines a series of SMS-related technologies such as FSE sequence, Hadamard pulse encoding, multi-band pulse, and quadratic phase modulation to effectively improve the signal-to-noise ratio of the low-field FSE sequence.

[0086] Compared with SMS technology based on multi-coil parallel acceleration under high-field magnetic resonance, the present invention does not rely on multi-coil technology. It can achieve multi-layer parallel excitation and imaging under single-coil conditions through the commonly used repeated scanning method under low field, and improve the imaging signal-to-noise ratio of related sequences; compared with simple repeated scanning under low field, the present invention makes more effective use of the time of each scan and effectively improves the overall signal-to-noise efficiency.

[0087] At present, the present invention has been applied to a self-built 110mT mobile scanning platform and has been proven to be effectively applied to conventional human brain scans in FSE sequences. In addition to improving the signal-to-noise ratio, the present invention can also be applied to accelerated imaging and has been proven to be applicable to multi-contrast imaging in FSE sequences, including T1-weighted (T1 weighted, T1W), T2-weighted (T2weighted, T2W), T1-weighted fluid attenuation (T1W FluidAttenuated Inversion Recovery, T1 FLAIR) and T2-weighted fluid attenuation (T2W FluidAttenuated Inversion Recovery, T2 FLAIR) and other contrasts. In the future, the present invention is expected to be more widely used in low-field magnetic resonance imaging, universally improving the signal-to-noise ratio of low-field imaging.

[0088] The preferred embodiments of the present invention have been described in detail above. It should be understood that numerous modifications and variations based on the concepts of the present invention are possible without inventive effort by those skilled in the art. Therefore, any technical solution that can be derived by one skilled in the art through logical analysis, reasoning, or limited experimentation based on the concepts of the present invention and the prior art should be within the scope of protection defined by the claims.

Claims

1. A single-coil multi-slice parallel magnetic resonance imaging method based on multiple excitations at low field, characterized in that: The following steps are involved: Step 1: using a multi-band pulse to achieve multi-layer parallel excitation, wherein the multi-band pulse is subjected to 0-π phase modulation; Step 2: construct a quadratic phase modulated fast spin echo sequence; Step 3, performing imaging using the quadratic phase modulated fast spin echo sequence; Step 4: Reconstruct the image by classifying the odd and even echoes to obtain the scan result image.

2. The single-coil multi-slice parallel magnetic resonance imaging method based on multiple excitations under low field as claimed in claim 1, characterized in that: The multi-band in step 1 is N bands, the phases of the N excitation pulses of the N bands correspond to the N-order Hadamard matrix, and N layers of images are acquired by N repeated excitations, where N is 4 or 8.

3. The single-coil multi-slice parallel magnetic resonance imaging method based on multiple excitations under low field as claimed in claim 2, characterized in that: The N frequency bands are 4 frequency bands, and the phase of the corresponding refocusing pulse is [0, π, π, 0].

4. The single-coil multi-slice parallel magnetic resonance imaging method based on multiple excitations under low field as claimed in claim 2, characterized in that: The N frequency bands are 8 frequency bands, and the phases of the corresponding refocusing pulses are [0, π, 0, π, π, 0, π, 0].

5. The single-coil multi-slice parallel magnetic resonance imaging method based on multiple excitations under low field as claimed in claim 2, characterized in that: The step 2 specifically includes the following steps: Step 2.1, applying an excitation pulse and a corresponding refocusing pulse within a repetition time; Step 2.2: Apply a layer selection gradient while applying each pulse; Step 2.3: Apply two phase encoding gradients of equal magnitude and opposite direction between the two refocusing pulses. Step 2.4: Apply a frequency encoding gradient between two phase encoding gradients while simultaneously acquiring the signal. The emission phase of each refocusing pulse and the receiving phase of the corresponding signal satisfy the quadratic phase modulation method; and within one repetition time, multiple groups of signal excitation and acquisition with different level combinations are performed.

6. The single-coil multi-slice parallel magnetic resonance imaging method based on multiple excitations under low field as claimed in claim 5, characterized in that: The step 2 further comprises the following steps: Add a destruction gradient before and after each layer selection gradient; At the end of each repetition time, a return pulse is added.

7. The single-coil multi-slice parallel magnetic resonance imaging method based on multiple excitations under low field as claimed in claim 5, characterized in that: The echo train length of the quadratic phase modulated fast spin echo sequence in step 2 is not less than 8.

8. The single-coil multi-slice parallel magnetic resonance imaging method based on multiple excitations under low field as claimed in claim 7, characterized in that: The transmit phase of each refocusing pulse of the quadratic phase modulated fast spin echo sequence in step 2 satisfies the quadratic phase modulation method. Specifically, the receive phase Φ is defined as rx0 is 0 degrees, the emission phase of the nth refocusing pulse Φ txn Satisfy Φ txn =Φ rx(n-1) +δ n , the receiving phase Φ when receiving the signal for the nth time rxn Satisfy Φ rxn =Φ rxn +2·δ n , where δ satisfies δ n =δ n-1 +Δ n , Δ n When n is less than or equal to 7, it is a specific value, that is, Δ1 = 68.92°, Δ2 = 69.35°, Δ3 = 81.22°, Δ4 = 71.15°, Δ5 = 46.67°, Δ6 = 71.16°, Δ7 = 101.55°, Δ n When n is greater than or equal to 8, the value is fixed at 68.92°.

9. The single-coil multi-slice parallel magnetic resonance imaging method based on multiple excitations under low field as claimed in claim 5, characterized in that: The step 3 specifically includes the following steps: For the N-layer parallel excitation imaging process, the scan uses the corresponding N excitation pulses and the corresponding refocusing pulses to perform 2N imaging times, that is, the same phase encoding gradient is used in two adjacent repetition times, but the phase encoding values between adjacent odd and even echoes are exchanged, and this process is repeated N times to obtain the complete k-space of odd and even echo images respectively.

10. The single-coil multi-slice parallel magnetic resonance imaging method based on multiple excitations under low field as claimed in claim 9, characterized in that: The step 4 specifically includes the following steps: Step 4.1, combining the data of all odd echoes into odd-echo k-space, and combining the data of all even echoes into even-echo k-space; Step 4.2: Decode the combined odd echo data set and even echo data set according to the Hadamard coding principle and reconstruct them according to the traditional image reconstruction process. The two sets of images are combined to obtain a scanning result image.

Images