An ultra-low field magnetic resonance imaging method based on planar echo imaging

Abstract

The application relates to the technical field of magnetic resonance imaging, in particular to a super-low-field magnetic resonance imaging method based on planar echo imaging, which comprises the following steps: S1, planar echo imaging is carried out by using a super-low-field magnetic resonance system to obtain echo data; S2, temperature drift phase change in the echo data is corrected according to coordinate coding to obtain correction data; S3, a de-artifact image and a magnetic field inhomogeneity map are respectively generated according to the correction data; and S4, the de-artifact image is corrected for image distortion by using the magnetic field inhomogeneity map, a target image is obtained and output. The beneficial effect is that coordinate coding is added in the planar echo imaging process, and then effective correction of temperature drift phase change in the echo data is realized based on the coordinate coding, so that the correction data are obtained. Then, a de-artifact image is obtained by respectively eliminating artifacts, and a magnetic field inhomogeneity map is calculated, and then the correction of image distortion is carried out based on the magnetic field inhomogeneity map, so that the target image is finally obtained and output.

Description

Technical Field

[0001] This invention relates to the field of magnetic resonance imaging technology, and more specifically to an ultra-low field magnetic resonance imaging method based on planar echo imaging. Background Technology

[0002] Magnetic resonance imaging (MRI) is a non-invasive imaging technique widely used in medicine, biology, and materials science. In recent years, ultra-low field MRI has attracted widespread attention both domestically and internationally due to its advantages such as good accessibility, ultra-quiet operation, no need for electromagnetic shielding, small size and lightweight design, and portability to the patient's bedside. Currently, ultra-low field MRI technology employs gradient echo (GRE) and fast spin echo acquisition techniques, successfully implementing several basic brain imaging protocols and providing clinically valuable information for the diagnosis of stroke and tumors.

[0003] In the existing technology, there are already technical solutions for imaging based on ultra-low field magnetic resonance imaging (MRI) systems. For example, Chinese patent CN201810511209.2 discloses a compressed sensing low-field MRI algorithm, which undersamples the acquired raw low-field MRI data to obtain random variable-density K-space data. The undersampled K-space data is then combined with compressed sensing theory to model the problem, which is a linear combination minimization problem containing data fidelity terms, sparse prior terms, and total variational terms. The dual-tree wavelet transform and wavelet tree sparse joint are used as the sparse transform in compressed sensing theory. The compressed sensing MRI algorithm using the dual-tree wavelet transform and wavelet tree sparse joint is used to solve this problem, obtaining the low-field MRI image reconstructed from the undersampled variable-density K-space data. Introducing the dual-tree wavelet transform as the sparse transform in the CS-MRI model overcomes the shortcomings of traditional wavelet transform, such as sensitivity, shift-variability, lack of directionality, and aliasing artifacts in the reconstructed image.

[0004] However, in actual implementation, the inventors found that the above-mentioned acquisition techniques based on gradient echo (GRE) and fast spin echo have long scanning times and are easily affected by motion, thus affecting image quality. For example, during the acquisition process, magnetic field temperature drift may occur, leading to a decrease in image quality. Summary of the Invention

[0005] To address the aforementioned problems in existing technologies, an ultra-low field magnetic resonance imaging method based on planar echo imaging is provided.

[0006] The specific technical solution is as follows:

[0007] An ultra-low field magnetic resonance imaging method based on planar echo imaging includes:

[0008] Step S1: Obtain echo data by performing planar echo imaging using an ultra-low field magnetic resonance system;

[0009] During planar echo imaging, coordinate encoding is added to the echo data through phase displacement;

[0010] Step S2: Correct the temperature drift phase change in the echo data according to the coordinate encoding in the echo data to obtain the correction data;

[0011] Step S3: Generate artifact-free images and magnetic field inhomogeneity maps according to the correction data;

[0012] Step S4: Use the magnetic field inhomogeneity map to correct image distortion in the artifact-free image, obtain the target image, and output it.

[0013] On the other hand, step S1 includes:

[0014] Step S11: Sequentially determine a TR value, and use the ultra-low field magnetic resonance system to perform imaging at the current TR value;

[0015] Step S12: The ultra-low field magnetic resonance system is used to sequentially output one of the pulse signals;

[0016] Step S13: Offset the TE value of the pulse signal to add the coordinate code, then return to step S12 until the preset scanning cycle is reached;

[0017] Step S14: Determine whether data collection is complete;

[0018] If so, proceed to step S2;

[0019] If not, return to step S11 to obtain the next TR value.

[0020] On the other hand, step S2 includes:

[0021] Step S21: The echo data is translated according to the coordinate encoding so that multiple frames of echo data have the same data phase with the same k-space center;

[0022] Step S22: Calculate the phase change value of a single excitation image from the echo data;

[0023] Step S23: Remove the phase change values ​​from the single-shot images to obtain the correction data.

[0024] On the other hand, the coordinate codes are divided into three groups, with each group of coordinate codes differing by a phase code line delay;

[0025] In step S21, the corresponding number of phase encoding line delays are removed according to the coordinate encoding added to the single excitation image to perform data phase alignment.

[0026] On the other hand, step S22 includes:

[0027] Step S221: For each of the k-space data of the single-shot image, calculate the data difference map corresponding to the k-space data of the first single-shot image;

[0028] Step S222: Summate each group of data difference maps along the non-phase encoding direction to obtain multiple one-dimensional vectors;

[0029] Step S223: For each group of one-dimensional vectors, search for the maximum value of the one-dimensional vector to obtain the maximum value position index;

[0030] Step S224: Determine the phase change according to the maximum value position index.

[0031] On the other hand, step S3 includes:

[0032] Step S31: The correction data is grouped according to the coordinate encoding, and then a complex average is performed to obtain a first complex average image and a second complex average image;

[0033] Step S32: Generate the artifact-removed image and the magnetic field inhomogeneity map based on the first complex average image and the second complex average image, respectively.

[0034] On the other hand, in step S32, the method for generating the artifact-free image includes:

[0035] Step A321: Summing the first complex average image and the second complex average image yields the artifact-free image.

[0036] On the other hand, in step S32, the method for generating the magnetic field non-uniformity map includes:

[0037] Step B321: Perform complex conjugate calculation on the first complex average image and the second complex average image to obtain a difference image;

[0038] Step B322: Extract the phase map from the difference map;

[0039] Step B323: Generate the magnetic field inhomogeneity map based on the time difference between the first complex average image and the second complex average image and the phase map.

[0040] On the other hand, step S4 includes:

[0041] Step S41: The artifact-removed image is masked according to the magnetic field inhomogeneity map to remove the region without magnetic resonance signal to obtain the masked image;

[0042] Step S42: Perform median filtering on the mask image to obtain the target image.

[0043] The above technical solution has the following advantages or beneficial effects:

[0044] To address the issue that existing low-field magnetic resonance imaging (MRI) technologies are susceptible to temperature drift and other factors due to long imaging times, this solution optimizes the imaging process for this scenario. Coordinate encoding is added during planar echo imaging, and this encoding is then used to effectively correct for temperature drift phase changes in the echo data, resulting in corrected data. Subsequently, artifact removal is performed to obtain a de-artifact map, and a magnetic field inhomogeneity map is calculated. Image distortion is then corrected based on the magnetic field inhomogeneity map, ultimately yielding the target image output. Attached Figure Description

[0045] Embodiments of the invention will be described more fully with reference to the accompanying drawings. However, the drawings are for illustration and explanation only and do not constitute a limitation on the scope of the invention.

[0046] Figure 1 This is an overall schematic diagram of an embodiment of the present invention;

[0047] Figure 2 This is a schematic diagram of sub-step S1 in an embodiment of the present invention;

[0048] Figure 3 This is a schematic diagram of sub-step S2 in an embodiment of the present invention;

[0049] Figure 4 This is a schematic diagram of sub-step S22 in an embodiment of the present invention;

[0050] Figure 5 This is a schematic diagram of sub-step S3 in an embodiment of the present invention;

[0051] Figure 6 This is a schematic diagram of sub-step S32 in an embodiment of the present invention;

[0052] Figure 7 This is a schematic diagram of sub-step S4 in an embodiment of the present invention. Detailed Implementation

[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0054] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0055] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but this is not intended to limit the scope of the invention.

[0056] This invention includes:

[0057] An ultra-low field magnetic resonance imaging method based on planar echo imaging, such as Figure 1 As shown, it includes:

[0058] Step S1: Obtain echo data by performing planar echo imaging using an ultra-low field magnetic resonance system;

[0059] In the process of planar echo imaging, coordinate encoding is added to the echo data through phase displacement;

[0060] Step S2: Correct the temperature drift phase change in the echo data according to the coordinate encoding in the echo data to obtain the correction data;

[0061] Step S3: Generate artifact-free images and magnetic field inhomogeneity maps based on the correction data;

[0062] Step S4: Use the magnetic field inhomogeneity map to correct image distortion in the artifact-free image, obtain the target image and output it.

[0063] Specifically, addressing the issue that existing low-field magnetic resonance imaging (MRI) technologies are susceptible to temperature drift and other factors due to long imaging times, this solution optimizes the imaging process for this scenario. During planar echo imaging, coordinate encoding is added through phase displacement. This coordinate encoding then effectively corrects for temperature drift phase changes in the echo data, resulting in calibrated data. Subsequently, artifact removal is performed to obtain an artifact-free image, and a magnetic field inhomogeneity map is calculated. Image distortion is then corrected based on the magnetic field inhomogeneity map, ultimately yielding the target image output. This process eliminates Nyquist artifacts and image distortion caused by magnetic field inhomogeneity and eddy current effects, reduces the impact of magnetic field temperature drift on the image, enables multi-contrast imaging, improves image quality, and reduces scanning time.

[0064] In practice, the aforementioned ultra-low field magnetic resonance imaging method is primarily configured as a software implementation in corresponding magnetic resonance imaging equipment, such as the 0.053T ultra-low field magnetic resonance imaging system produced by Hangzhou Microfilm Medical Technology Co., Ltd. Echo data, correction data, etc., refer to planar echo imaging (EPI) data obtained from a single excitation using this type of magnetic resonance imaging equipment.

[0065] In one embodiment, such as Figure 2As shown, step S1 includes:

[0066] Step S11: Determine a TR value sequentially, and use an ultra-low field magnetic resonance imaging system to perform imaging at the current TR value;

[0067] Step S12: Use an ultra-low field magnetic resonance system to output a pulse signal sequentially;

[0068] Step S13: Offset the TE value of the pulse signal to add coordinate encoding, then return to step S12 until the preset scanning cycle is reached;

[0069] Step S14: Determine whether data collection is complete;

[0070] If so, proceed to step S2;

[0071] If not, return to step S11 to obtain the next TR value.

[0072] Specifically, to implement the process of adding phase encoding to the planar echo image of a single excitation, in this embodiment, after determining the TR value of the low-field magnetic resonance system during imaging, single-excitation imaging is performed. During each single-excitation imaging process, the TE value is shifted to obtain single-excitation images with different TE values ​​under the same TR value as echo data. This process is then repeated until echo data acquisition is complete. In one embodiment, the TE values ​​under the same group of TR values ​​include three; that is, during the single-excitation imaging process under every three adjacent TR values, a phase encoding is shifted to obtain three single-excitation images with different TE values. The time ΔT of a phase encoding line between the single-excitation images in the same group is defined, and TE1... <TE2<TE3。

[0073] In one embodiment, such as Figure 3 As shown, step S2 includes:

[0074] Step S21: The echo data is translated according to coordinate encoding so that multiple frames of echo data have the same data phase at the same k-space center;

[0075] Step S22: Calculate the phase change value of a single excitation image from the echo data;

[0076] Step S23: Remove the phase change values ​​from the single-shot images to obtain the correction data.

[0077] Specifically, to achieve a favorable temperature drift correction process, in this embodiment, after the aforementioned phase encoding imaging process is introduced, the acquired echo data is shifted according to the TE value at the time of its acquisition. For example, in one embodiment, there are three sets of coordinate encodings, and each set of coordinate encodings differs by one phase encoding line delay, so that during the acquisition process, the same set of single-shot images including three different TE values can be obtained, denoted as TE1, TE2, TE3, and TE1 < TE2 < TE3; then in step S21, the middle single-shot image TE2 is taken as the reference image, and the single-shot images TE1 and TE3 at both ends are respectively shifted to align with the reference image TE2, that is, the single-shot image TE1 is shifted up by one phase encoding line, and the single-shot image TE3 is shifted down by one phase encoding line. Through the aforementioned cyclic shifting operation, multiple frames of echo data are made to have the same data phase at the k-space center. Subsequently, the phase variation caused by magnetic field temperature drift of each single-shot EPI image is corrected, so as to obtain corrected data and achieve a favorable correction effect.

[0078] In one embodiment, as shown in Figure 4 , step S22 includes:

[0079] Step S221: For the k-space data of each single-shot image, respectively calculate a data difference map corresponding to the k-space data of the first single-shot image;

[0080] Step S222: Sum each set of data difference maps along the non-phase encoding direction respectively to obtain a plurality of one-dimensional vectors;

[0081] Step S223: For each set of one-dimensional vectors, respectively search for the maximum value of the one-dimensional vector to obtain the maximum value position index;

[0082] Step S224: Determine the phase variation value according to the maximum value position index.

[0083] Specifically, to realize accurate judgment of the phase variation value caused by temperature drift, in this embodiment, for the k-space data k1, k2, ..., kn of each single-shot image, the data difference map corresponding to the k-space data of the first single-shot image is calculated respectively. Subsequently, the data difference maps are summed along the non-phase encoding direction to obtain a plurality of 1×Ny one-dimensional vectors, where Ny is the number of phase encoding steps. Search for the position index y1, y2, ......y of the maximum value of each one-dimensional vector n , and convert it into phase variation, that is, linearly increase the range from -pi to pi according to the number of phase encoding steps to obtain Ny elements, and then multiply by the corresponding maximum value position index to obtain the corresponding phase variations Ph1, Ph2, ......, Ph nThe above process enables effective calculation of phase changes caused by temperature drift. Based on the above calculation process, in step S23, the k-space data of each single excitation EPI can be calculated according to formula k i *e -1j*Ph i To compensate for the phase change caused by magnetic field temperature drift, where i corresponds to the EPI data excited by the i-th word, and j is the imaginary unit of the complex number.

[0084] In one embodiment, such as Figure 5 As shown, step S3 includes:

[0085] Step S31: Group the correction data according to coordinate encoding, and then perform complex averaging to obtain the first complex average image and the second complex average image;

[0086] Step S32: Generate an artifact-free image and a magnetic field inhomogeneity map based on the first complex average image and the second complex average image, respectively.

[0087] Specifically, after effectively correcting the phase change caused by temperature drift based on the above process, in this embodiment, the correction data is grouped according to coordinate encoding, and then a complex average is performed to obtain a first complex average image and a second complex average image. The first complex average image and the second complex average image are processed respectively to remove artifacts from the image and calculate the magnetic field inhomogeneity.

[0088] Specifically, such as Figure 6 As shown:

[0089] In step S32, the method for generating the artifact-free image includes:

[0090] Step A321: Summing the first complex average image and the second complex average image yields the artifact-free image.

[0091] Specifically, in order to effectively remove Nyquist artifacts from an image, in this embodiment, after performing the above-mentioned complex averaging calculation process, the first complex average image and the second complex average image are summed to obtain the artifact-free image.

[0092] Furthermore, in step S32, the method for generating the magnetic field non-uniformity map includes:

[0093] Step B321: Perform complex conjugate calculation on the first complex average image and the second complex average image to obtain the difference image;

[0094] Step B322: Extract the phase map from the difference map;

[0095] Step B323: Generate a magnetic field inhomogeneity map based on the time difference and phase map of the first complex average image and the second complex average image.

[0096] Specifically, to calculate the magnetic field inhomogeneity, in this embodiment, after performing the aforementioned complex averaging calculation, a difference map is obtained by performing a complex conjugate calculation on the first and second complex average images. Subsequently, the phase map is extracted from the difference map, and then the phase map is divided by the time difference between the first and second complex average images to calculate the magnetic field inhomogeneity.

[0097] In one embodiment, such as Figure 7 As shown, step S4 includes:

[0098] Step S41: Perform masking on the magnetic field inhomogeneity map to remove areas without magnetic resonance signals and obtain a masked image;

[0099] Step S42: Perform median filtering on the mask image to obtain the magnetic field correction image;

[0100] Step S43: Obtain the image by performing distortion correction on the artifact-free image based on the magnetic field correction image.

[0101] Specifically, to correct image distortion, the image of magnetic field inhomogeneity is first masked based on its amplitude map to remove regions without magnetic resonance signals. Then, a 2D median filter (5×5) is used for smoothing, thereby improving the signal-to-noise ratio of the image. The following formula is mainly used in the subsequent image distortion correction process:

[0102]

[0103] Where S(x,k) y I(x,y) is the target image obtained after correction in the phase encoding direction, I(x,y) is the artifact-free image, and N is the target image obtained after correction in the phase encoding direction. y ΔT represents the total number of phase codes, ΔT represents the time difference between the phase code lines, and Bmap(x,y) represents the magnetic field correction image.

[0104] The above are merely preferred embodiments of the present invention and are not intended to limit the implementation methods and protection scope of the present invention. Those skilled in the art should recognize that any equivalent substitutions and obvious changes made based on the description and illustrations of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for ultra-low field magnetic resonance imaging based on planar echo imaging, characterized in that, include: Step S1: Obtain echo data by performing planar echo imaging using an ultra-low field magnetic resonance system; During planar echo imaging, coordinate encoding is added to the echo data through phase displacement; Step S2: Correct the temperature drift phase change in the echo data according to the coordinate encoding in the echo data to obtain the correction data; Step S3: Generate artifact-free images and magnetic field inhomogeneity maps according to the correction data; Step S4: Use the magnetic field inhomogeneity map to correct image distortion in the artifact-removed image, obtain the target image, and output it; Step S2 includes: Step S21: The echo data is translated according to the coordinate encoding so that multiple frames of echo data have the same data phase with the same k-space center; Step S22: Calculate the phase change value of a single excitation image from the echo data; Step S23: Remove the phase change values ​​from the single-shot images to obtain the correction data; The coordinate codes are divided into three groups, with each group of coordinate codes differing by a phase code line delay. In step S21, the single-shot image located in the middle is used as a reference image, and the single-shot images at both ends are translated to align with the reference image. Step S22 includes: Step S221: For each of the k-space data of the single-shot image, calculate the data difference map corresponding to the k-space data of the first single-shot image; Step S222: Summate each group of data difference maps along the non-phase encoding direction to obtain multiple one-dimensional vectors; Step S223: For each group of one-dimensional vectors, search for the maximum value of the one-dimensional vector to obtain the maximum value position index; Step S224: Determine the phase change value according to the maximum value position index.

2. The ultra-low field magnetic resonance imaging method according to claim 1, characterized in that, Step S1 includes: Step S11: Sequentially determine a TR value, and use the ultra-low field magnetic resonance system to perform imaging at the current TR value; Step S12: The ultra-low field magnetic resonance system is used to output a pulse signal sequentially; Step S13: Offset the TE value of the pulse signal to add the coordinate code, then return to step S12 until the preset scanning cycle is reached; Step S14: Determine whether data collection is complete; If so, proceed to step S2; If not, return to step S11 to obtain the next TR value.

3. The ultra-low field magnetic resonance imaging method according to claim 1, characterized in that, Step S3 includes: Step S31: The correction data is grouped according to the coordinate encoding, and then a complex average is performed to obtain a first complex average image and a second complex average image; Step S32: Generate the artifact-removed image and the magnetic field inhomogeneity map based on the first complex average image and the second complex average image, respectively.

4. The ultra-low field magnetic resonance imaging method according to claim 3, characterized in that, In step S32, the method for generating the artifact-free image includes: Step A321: Summing the first complex average image and the second complex average image yields the artifact-free image.

5. The ultra-low field magnetic resonance imaging method according to claim 3, characterized in that, In step S32, the method for generating the magnetic field non-uniformity map includes: Step B321: Perform complex conjugate calculation on the first complex average image and the second complex average image to obtain a difference image; Step B322: Extract the phase map from the difference map; Step B323: Generate the magnetic field inhomogeneity map based on the time difference between the first complex average image and the second complex average image and the phase map.

6. The ultra-low field magnetic resonance imaging method according to claim 1, characterized in that, Step S4 includes: Step S41: The artifact-removed image is masked according to the magnetic field inhomogeneity map to remove the region without magnetic resonance signal to obtain the masked image; Step S42: Perform median filtering on the mask image to obtain the target image.

Images