Parallel PROPELLER MR Imaging

Abstract

This invention relates to a method for performing MR imaging on a patient's body (10). The object of this invention is to provide a method for reducing artifact levels in parallel (accelerated) PROPELLER imaging. According to the proposed invention, MR signals are acquired in parallel via multiple RF receiving antennas (11, 12, 13) with different spatial sensitivity distributions, wherein the MR signals are acquired alternately from multiple k-space blades (21-26) according to a PROPELLER scheme. The k-space blades (21-26) rotate around the center of the k-space, wherein the total set of acquired MR signals spans a circle (28) in the k-space, and the common central region (27) of the k-space is covered by all the k-space blades (21-26). The acquisition of MR signals from the respective k-space blades (21-26) involves subsampling of the k-space, wherein the relative k-space sampling density varies with the rotation angle. Finally, an MR image is computed by combining the MR signals acquired via the different RF receiving antennas (11, 12, 13) using their spatial sensitivity distributions to perform parallel image reconstruction. Furthermore, the present invention relates to an MR system (1) and a computer program for the MR system (1).

Description

Technical Field

[0001] This invention relates to the field of magnetic resonance (MR) imaging. It relates to a method for MR imaging of a patient's body placed within the examination volume of an MR system. The invention also relates to an MR system and a computer program to run on the MR system. Background Technology

[0002] Image-forming MR methods, which utilize the interaction between magnetic fields and nuclear spins to form two-dimensional or three-dimensional images, are now widely used, especially in the field of medical diagnostics, because they are superior to other imaging methods in many ways for soft tissue imaging, do not require ionizing radiation, and are generally non-invasive.

[0003] The so-called periodically rotating overlapping parallel lines with enhanced reconstructed imaging (PROPELLER) offer advantages for clinical MR imaging, such as robustness to patient motion and inherent motion compensation capabilities. In the PROPELLER concept (see Pipe et al., Magnetic Resonance in Medicine, 42: 963-969, 1999), MR signal data is acquired in k-space in N strips, each strip comprising L parallel k-space lines corresponding to the L lowest-frequency phase-encoded lines in a Cartesian-based k-space sampling scheme. Each strip (referred to herein as a k-space blade) is rotated in k-space by, for example, an angle of 180° / N, such that the total collection of MR signals spans a circle in k-space. If a full k-space data matrix with a diameter M is desired, L and N can be chosen such that… L × N = M ×π / 2. A fundamental characteristic of conventional PROPELLER is the acquisition of a central circular portion with diameter L in k-space for each k-space leaf. This central portion can be used to reconstruct low-resolution MR images of each k-space leaf. These low-resolution MR images can be compared to each other to estimate and compensate for patient motion. For example, in-plane translations and rotations between k-space leaves can be removed in this way. Therefore, the PROPELLER technique utilizes oversampling in the central portion of k-space to obtain a motion-robust MR image acquisition technique relative to the patient being examined during MR signal acquisition. Furthermore, due to the averaging of the k-space leaves in the PROPELLER, different types of imaging artifacts are "averaged out" to some extent.

[0004] A major drawback associated with PROPELLER is the prolonged scan time. Various studies have explored applying parallel imaging acceleration to PROPELLER as a means of mitigating this limitation (see Chang et al., MagneticResonance in Medicine, 74:1598-1608, 2015). Parallel imaging acceleration has the potential to widen the k-space blades (because the sampling density of each k-space blade in the phase-encoding direction is reduced), thereby reducing scan time by requiring fewer blades to cover the entire k-space. Nevertheless, noise and artifacts are problems in accelerating PROPELLER, especially at high acceleration rates. The levels of noise and artifacts affect not only image quality but also the accuracy of motion correction. In common reconstruction strategies used to accelerate PROPELLER, parallel image reconstruction is used to reconstruct each k-space blade individually, and then all single-blade images are combined with motion correction. While the effectiveness of this approach has been convincingly demonstrated at relatively low acceleration rates and at higher acceleration rates, certain limitations of this approach become apparent, particularly regarding subsampling artifacts and noise amplification. As a result, the tuning of parallel image reconstruction can be highly dependent on the orientation of the individual k-space blades. This results in noticeable artifacts and noise. A small number of poorly calibrated leaves with similar orientations can significantly increase the level of noise and artifacts in the final (combined) MR image. This problem has been observed to persist even when all leaves are jointly reconstructed into the final MR image in a single step to better utilize the complementary information of different leaves, as proposed by Lyu et al. (ScientificReports 2017, 7:42602).

[0005] US 2023 / 0324487 A1 discloses a method for accelerating magnetic resonance imaging, comprising applying a pulse sequence by an MR system having k-space trajectories of multiple blades rotating in k-space, each blade including multiple views, wherein the k-space trajectories have an undersampling pattern in k-space. The method also includes receiving k-space data of an object acquired by the pulse sequence, reconstructing an MR image of the object based on the k-space data using compressed sensing, and outputting the reconstructed image. Considering blades of fixed width, it further discloses a fixed number of acquired k-space lines differently distributed on a fixed number of nominal k-space lines in each blade.

[0006] Larson PEZ et al.'s paper, "Anisotropic field-of-view shapes for improved PROPELLER imaging" (Magnetic Resonance Imaging, Vol. 27, No. 4, May 4, 2009, pp. 470-479, DOI: 10.1016 / J.MRl.2008.07.023), proposes a novel algorithm for customizing PROPELLER acquisition to flexible and accurate desired field-of-view (FOV) shapes and sizes. FOV design also allows for better motion correction and reduced aliasing artifacts in rotational motion. Some possible FOV shapes include arcuate ellipses, ovals, and rectangles, and any convex π-symmetric shape can be designed. Standard PROPELLER reconstruction was used with minor modifications, and results with simulated motion confirm the effectiveness of motion correction with these modified FOV shapes. Summary of the Invention

[0007] As can be readily understood from the foregoing, an improved PROPELLER MR imaging technique is needed. The object of this invention is to provide a method for reducing the level of artifacts in parallel (accelerated) PROPELLER imaging.

[0008] According to the present invention, a method for performing MR imaging on a patient's body placed within the examination volume of an MR system is disclosed. The method includes the following steps:

[0009] MR signals are generated by subjecting at least a portion of the body to an MR imaging sequence comprising multiple radio frequency (RF) pulses and switched magnetic field gradients.

[0010] The MR signals are acquired in parallel via multiple RF receiving antennas with different spatial sensitivity distributions. The MR signals are acquired alternately from multiple k-space blades according to the PROPELLER scheme, with each k-space blade rotating around the center of the k-space. The total acquired MR signal set spans a circle within the k-space, and the common central region of the k-space is covered by all k-space blades. MR signal acquisition from each corresponding k-space blade involves sub-sampling within the k-space, and the relative k-space sampling density varies with the rotation angle.

[0011] MR images are calculated by combining the MR signals acquired via the different RF receiving antennas using the spatial sensitivity distribution of the different RF receiving antennas to perform parallel image reconstruction.

[0012] In the context of this invention, the term relative k-space sampling density that varies with rotation angle is equivalent to the acceleration factor that varies with rotation angle.

[0013] US 2023 / 0324487, considering blades of fixed width, further discloses a fixed number of k-space lines in each blade, with each blade having a fixed number of collected k-space lines distributed differently.

[0014] Larson PEZ et al. proposed a strategy for designing PROPELLER acquisitions, in the most general case where the number of k-space lines acquired per blade varies, as does the spacing between those k-space lines. This strategy does not consider parallel imaging to suppress aliasing from undersampling, and therefore does not consider acceleration factors. For any non-square or non-circular FOV, the absolute nominal k-space sampling density differs in both directions. Therefore, even for PROPELLER acquisitions with an elliptical FOV but without parallel imaging, the absolute nominal k-space sampling density will vary depending on the blade's rotation angle. Thus, the different acceleration factors for different blades should be characterized by the density of the actual k-space sampling relative to the density of the nominal k-space sampling required to cover and resolve a given field of view at a specified spatial resolution.

[0015] This invention proposes acquiring k-space blades using a known PROPELLER scheme combined with parallel imaging. To accelerate MR signal acquisition via parallel imaging, k-space subsampling is applied. This invention proposes varying the degree of subsampling (in the phase-encoding direction of the Cartesian sampling of an individual k-space blade) according to the rotation angle of the corresponding k-space blade. Changing the sampling density means that the acceleration factor varies from k-space blade to k-space blade. Therefore, k-space blades differ from each other in terms of acceleration factor, i.e., the distance between adjacent actually acquired (parallel) k-space lines differs from the distance between adjacent nominally acquired (parallel) k-space lines (without parallel imaging acceleration). This allows for the adaptation of subsampling for each k-space blade to the RF receiving antenna (RF coil) employed to support parallel image reconstruction for the corresponding orientation of the k-space blade. Therefore, artifacts can be reduced in the final MR image.

[0016] Within the scope of this invention, MR signal acquisition without subsampling can be applied to some rotation angles. However, at least for a subset of k-space blades, subsampling must be applied to achieve acceleration.

[0017] In this embodiment, all k-space blades comprise the same number of parallel and equidistant k-space lines, but have different widths in their respective phase encoding directions. Changing the width of the k-space blades is one possible way to alter the (uniform) k-space sampling density based on the rotation angle of the respective k-space blades. In the case of anisotropic k-space resolution, the acceleration factor depends not only on the width of the k-space blades in the respective phase encoding direction, but also on the desired k-space resolution in that direction. Therefore, in principle, the (uniform) k-space sampling density can also be changed without altering the actual width of the k-space blades, which is also within the scope of this invention.

[0018] In a possible embodiment, parallel image reconstruction is performed according to the SENSE reconstruction method (see Pruessman et al., Magnetic Resonance in Medicine, 42:952-962, 1999). This can be combined with compressed sensing (CS) reconstruction methods (see Lustig et al., Magnetic Resonance in Medicine, 58:1182-1195, 2007). However, other known parallel imaging techniques can also be combined with the method of this invention. In SENSE, the spatial sensitivity distribution of the RF receiving antenna used is estimated during the step of calculating the MR image. These estimates can be obtained, for example, from corresponding pre-scans known in the art.

[0019] In a possible embodiment, the variation in the width of the k-space blades, depending on the rotation angle, is determined by the (essentially) elliptical shape of the common central region of the k-space. The long parallel edges of each rectangular k-space blade (parallel to each other and parallel to the k-space lines) are tangent to the common central region of the k-space. Due to the elliptical shape, the k-space blade whose k-space lines are parallel to the semi-major axis of the ellipse has the minimum width, i.e., the blade with the densest uniform k-space sampling, while the k-space blade whose k-space lines are parallel to the semi-minor axis of the ellipse has the maximum width, i.e., the lowest uniform sampling density. Therefore, the directions of the highest and lowest sampling densities are perpendicular to each other. The ratio of the highest to the lowest width of the k-space blade is defined by the ratio of the semi-major and semi-minor axes of the elliptical shape. Furthermore, the elliptical shape ensures that the width of the k-space blades varies continuously with the rotation angle, from a minimum value in the direction parallel to the semi-major axis to a maximum value in the direction perpendicular to the semi-major axis (i.e., parallel to the semi-minor axis). The term "elliptical shape" used herein is intended to exclude circular shapes, although in a strictly mathematical sense, a circle is a special case of an ellipse (with zero eccentricity). In a possible embodiment, the eccentricity of the elliptical shape is greater than 0.1, preferably greater than 0.2, more preferably greater than 0.5, and most preferably greater than 0.7. The eccentricity is a parameter that can be tuned, for example, by the operator to achieve the desired image quality (in terms of the level of artifacts and noise).

[0020] In another embodiment, the semi-minor axis of the elliptical shape is oriented in a direction where the difference in spatial sensitivity distribution between the RF receiving antennas is less pronounced than in a direction perpendicular to it. In other words, this means that the orientation of the semi-major and semi-minor axes of the elliptical shape is chosen, for example, by the operator, such that the sampling density is highest for the k-space blade with the worst conditioning for parallel image reconstruction. Small differences in spatial sensitivity lead to poor conditioning for parallel image reconstruction. According to the invention, this can be considered by selecting an appropriate orientation and eccentricity of the elliptical shape in the image plane, such that the employed RF antenna supports parallel image reconstruction for both the worst-case and best-case blade orientations. Simultaneously, the invention enables a good trade-off between image quality and acquisition speed.

[0021] In yet another embodiment, the k-space blade with the minimum k-space sampling density is oriented in a direction in which the patient's body extends furthest from the center of the examination volume of the MR system. Parallel imaging reconstruction is particularly affected by inaccurate estimations of the spatial sensitivity distribution of the RF receiving antenna used in this direction. Similarly, according to the invention, this can be taken into account by selecting appropriate orientations for the k-space blades with the maximum and minimum k-space sampling densities, respectively.

[0022] In conventional PROPELLER imaging, the method of the present invention may further include steps of estimating and compensating for patient motion. For example, low-resolution MR images reconstructed based on a common central region of k-space are compared to each other to remove in-plane translations and rotations between k-space blades. These factors should be corrected in each k-space blade according to the present invention before further processing of the MR signal. This makes the method of the present invention robust to motion of the patient being examined during MR signal acquisition.

[0023] The method of the present invention described so far can be performed by means of an MR system comprising: at least one main magnet coil for generating a uniform, stable magnetic field B0 within an examination volume; a plurality of gradient coils for generating switched magnetic field gradients in different spatial directions within the examination volume; at least one volume RF coil for generating RF pulses within the examination volume; a plurality of RF receiving antennas having different spatial sensitivity distributions for receiving MR signals from the body of a patient positioned within the examination volume; a control unit for controlling the timing alternation of the RF pulses and the switched magnetic field gradients; and a reconstruction unit for reconstructing MR images based on the received MR signals. The method of the present invention can be implemented by corresponding programming of the reconstruction unit and / or control unit of the MR system. (Parallel) image reconstruction can also be performed on a separate (remote) computer that is not a part of the MR system.

[0024] The method of the present invention can be advantageously performed on most MR systems currently used in clinical practice. For this purpose, only a computer program is needed to control the MR system, causing it to perform the steps of the method described above. The computer program can reside on a data carrier or in a data network so that it can be downloaded and installed in the control and / or reconstruction unit of the MR system. Attached Figure Description

[0025] The accompanying drawings disclose preferred embodiments of the invention. However, it should be understood that the drawings are for illustrative purposes only and are not intended to limit the scope of the invention. In the drawings:

[0026] Figure 1 An MR system for performing the method of the present invention is shown;

[0027] Figure 2 The modified PROPELLER acquisition scheme of the present invention is illustrated schematically. Detailed Implementation

[0028] refer to Figure 1The MR system 1 is shown as a block diagram. The system includes a superconducting or resistive master magnet coil 2, which generates a substantially uniform, time-constant master magnetic field B0 along the z-axis through the examination volume. The system also includes a set of shimming coils 2' (first-order, second-order, and, where applicable, third-order), wherein the current flowing through the individual shimming coils of the set 2' is controllable for the purpose of minimizing deviations in B0 within the examination volume.

[0029] Magnetic resonance generation and manipulation systems apply a series of RF pulses and switched magnetic field gradients to excite, reverse, or saturate nuclear magnetic spins to sense, refocus, or otherwise manipulate magnetic resonances to spatially or otherwise encode magnetic resonances, thereby performing MR imaging.

[0030] More specifically, gradient amplifier 3 applies current pulses or waveforms along the x, y, and z axes of the examination volume to selected whole-body gradient coils 4, 5, and 6. Digital RF frequency transmitter 7 transmits RF pulses or pulse packets to body RF coil 9 via transmit / receive switch 8 to deliver RF pulses into the examination volume. Typical MR imaging sequences consist of packets of short-duration RF pulse segments, which, together with any applied magnetic field gradient, enable the manipulation of magnetic resonance selection. RF pulses are also used to select portions of body 10 located within the examination volume. MR signals are also picked up by body RF coil 9.

[0031] To generate MR images of a limited area of ​​body 10 and to accelerate scanning via parallel imaging, a set of local array RF antennas (coils) 11, 12, and 13 are positioned adjacent to the area selected for imaging. The array coils 11, 12, and 13 can be used to receive MR signals induced by RF transmissions via the body RF coils.

[0032] The resulting MR signal is picked up by the bulk RF coil 9 and / or by the array RF coils 11, 12, 13, and demodulated by a receiver 14, which preferably includes a preamplifier (not shown). The receiver 14 is connected to the RF coils 9, 11, 12, and 13 via a transmit / receive switch 8.

[0033] The host computer 15 controls the shimming coil 2', gradient pulse amplifier 3, and transmitter 7 to generate any of a plurality of MR imaging sequences, such as echo plane imaging (EPI), echo volume imaging, gradient and spin echo imaging, fast spin echo imaging, etc. For the selected sequence, receiver 14 rapidly and continuously receives single or multiple MR data lines after each RF excitation pulse. Data acquisition system 16 performs analog-to-digital conversion of the received signals and converts each MR data line into a digital format suitable for further processing. In modern MR systems, data acquisition system 16 is a separate computer dedicated to acquiring raw image data.

[0034] Ultimately, the raw digital image data is reconstructed into an image representation by reconstruction processor 17, which applies Fourier transform or other suitable reconstruction algorithms, such as SENSE. MR images can represent planar slices across the patient, arrays of parallel planar slices, three-dimensional volumes, etc. The image is then stored in an image memory, which can be accessed, for example via a video monitor 18, to convert the slices, projections, or other portions of the image representation into a suitable format for visualization. The video monitor 18 provides a human-readable display of the resulting MR image.

[0035] The host computer 15 and the reconstruction processor 17 are programmed to perform the methods of the present invention as described above and below.

[0036] Figure 2 The illustration shows k-space sampling according to the adapted PROPELLER MR imaging method of the present invention. For example... Figure 2 As shown, six k-space blades 21-26 are acquired. Each blade 21-26 covers a different portion of the k-space, with a central elliptical region 27 of the k-space acquired for each blade 21-26. The blades 21-26 are rotated around the center of the k-space such that the total acquired MR dataset spans a circle 28 within the k-space. Each of the rectangular k-space blades 21-26 comprises 11 parallel and equidistant k-space lines of equal length, such that each k-space blade 21-26 provides uniform sampling of the rectangular region covered by the respective blade 21-26. In-plane translations and rotations between the blades 21-26 are detected and corrected using, for example, known similarity measurement methods as in conventional PROPELLER imaging.

[0037] The proposed PROPELLER scheme, combined with parallel imaging, means that at least some of the k-space blades 21-26 are subsampled in directions perpendicular to parallel and equidistant k-space lines of each blade (i.e., in the phase-encoding direction of the corresponding Cartesian k-space sampling pattern). In this case, the k-space resolution is isotropic, and the blades are uniformly subsampled in the phase-encoding direction of the corresponding Cartesian k-space sampling pattern. Figure 2As shown, k-space blades 21-26 have different widths in their respective phase coding directions (i.e., transverse to the k-space line / transverse to the longitudinal extension of the blade). K-space blade 21 is the widest blade. Therefore, this blade is assigned the lowest sampling density and the highest acceleration factor. K-space blade 26 is the narrowest blade. Therefore, this blade is assigned the highest sampling density and the lowest acceleration factor. The intermediate values ​​of sampling density and acceleration factor are attributed to the remaining k-space blades 22-25. The sampling density increases continuously from the widest k-space blade 21 to the narrowest k-space blade 26. The variation in width of k-space blades 21-26 from blade to blade allows for the adaptation of uniform subsampling of each k-space blade 21-26 to the capabilities of the employed RF receiving antennas 11, 12, and 13 to support parallel image reconstruction for the corresponding orientations of k-space blades 21-26.

[0038] from Figure 2 It can also be seen that the width of the k-space blades 21-26 is determined by the elliptical shape of the common central region 27 of the k-space, depending on the rotation angle. The long parallel edge of each rectangular k-space blade 21-26 is tangent to the common central region 27 of the k-space. Due to the elliptical shape, its k-space lines are parallel to the semi-major axis of the elliptical region 27 (in...). Figure 2 The k-space blade 24 (with a mid-horizontal orientation) provides the highest sampling density in the phase encoding direction, while its k-space line is parallel to the semi-minor axis of the elliptical region 27 (in... Figure 2 The k-space blade 21 (vertically oriented) provides the lowest sampling density. Therefore, the directions of the highest and lowest sampling densities are perpendicular to each other.

[0039] Each pair of adjacent k-space blades 21-26 shares two of their vertices on the circle 28 in the k-space traversed by the set of MR signals. This is in Figure 2 The diagram is illustrated by two small circles. These circles indicate vertices shared by k-space blades 21 and 22. These shared vertices are diagonally opposite each other on each blade. This is in... Figure 2 The diagram only shows k-space blades 21 and 22, but applies to all k-space blades 21-26. All adjacent k-space blades 21-26 are connected to each other at the outer circumference of a circle 28 having its vertices. Therefore, k-space sampling is performed without overlap or gaps at the circumference of circle 28. Thus, the rotation angle increments between adjacent k-space blades 21-26 vary. Specifically, in Figure 2 In this model, adjacent k-space blades 21 and 22 define angles between them that differ from those of k-space blades 22 and 23, which in turn define angles that differ from those of k-space blades 23 and 24. This contrasts with conventional PROPELLER imaging, which applies constant increments of rotation angles.

[0040] Based on the imaged anatomical structures and the spatial sensitivity distribution of the RF receiving antennas 11, 12, and 13 used, the parameters (lengths and orientations of the semi-major and semi-minor axes in k-space) of the elliptical region 27 can be set according to the requirements of the corresponding imaging task in terms of artifact and noise levels. In practical applications, the semi-minor axis of the elliptical shape should be oriented in a direction less pronounced than in a direction perpendicular to it, given that variations in the spatial sensitivity distribution of the RF receiving antennas 11, 12, and 13 are less significant. This means, in other words, that parameters should be set such that the (one or more) k-space blades 21-26 (which are the least favorable for parallel image reconstruction) are... Figure 2 (In the example of the vertical direction), the sampling density is the highest.

[0041] Sagittal and coronal knee imaging are illustrative applications of the proposed adapted PROPELLER technique. Typical RF coils for knee imaging support SENSE reconstruction well in the lateral and anteroposterior directions, but not so well in the pedicle direction. This is because the major axis of the coil element is aligned with the pedicle direction, thus providing only limited sensitivity variation in that direction. In Cartesian MRI, this is less critical because the readout (frequency encoding) direction can be selected in the pedicle direction, preventing subsampling in that direction. The method of the present invention can be applied to this example, wherein the semi-minor axis of the elliptical central region 27 is oriented in the pedicle direction to enable PROPELLER imaging of the knee in conjunction with SENSE at a reduced level of artifacts.

Claims

1. A method for performing MR imaging on the body (10) of a patient placed in the examination volume of a magnetic resonance MR system (1), the method comprising: MR signals are generated by subjecting at least a portion of the body (10) to an MR imaging sequence comprising multiple radio frequency (RF) pulses and switched magnetic field gradients. The MR signals are acquired in parallel via multiple RF receiving antennas (11, 12, 13) with different spatial sensitivity distributions, wherein the MR signals are acquired alternately over time from multiple k-space blades (21-26) according to the PROPELLER scheme, wherein the k-space blades (21-26) rotate around the center of the k-space, wherein the total acquired MR signal set spans a circle (28) in the k-space and the common central region (27) of the k-space is covered by all k-space blades (21-26), the acquisition of MR signals from the respective k-space blades (21-26) involves subsampling of the k-space, wherein the acceleration factor associated with the subsampling varies with the rotation angle; and MR images are calculated by combining the spatial sensitivity distributions of different RF receiving antennas (11, 12, 13) with the MR signals acquired via the different RF receiving antennas to perform parallel image reconstruction.

2. The method according to claim 1, wherein, The parallel image reconstruction is performed either independently according to the SENSE reconstruction method or according to the SENSE reconstruction method in combination with the compressed sensing reconstruction method.

3. The method according to claim 1 or 2, wherein, Each of the k-space blades (21-26) comprises the same number of parallel, equidistant k-space lines.

4. The method according to any one of claims 1 to 3, wherein, A k-space blade (24) with the minimum relative k-space sampling density is oriented perpendicular to another k-space blade (21) with the maximum relative k-space sampling density.

5. The method according to any one of claims 1 to 4, wherein, The rotation angle increments between the continuously acquired k-space blades are variable (21-26).

6. The method according to any one of claims 1 to 5, wherein, The width of the k-space blades (21-26) varies according to the rotation angle.

7. The method according to claim 6, wherein, The variation in the width of the k-space blades (21-26) depending on the rotation angle is determined by the elliptical shape of the common central region (27) of the k-space.

8. The method according to claim 7, wherein, The eccentricity of the elliptical shape is greater than 0.1, preferably greater than 0.2, more preferably greater than 0.5, and most preferably greater than 0.

7.

9. The method according to any one of claims 1 to 8, wherein, The k-space blade with the minimum k-space sampling density is oriented in the direction in which the difference between the spatial sensitivity distributions of the RF receiving antennas (11, 12, 13) is less pronounced than in the direction perpendicular to the direction, and the k-space blade with the maximum k-space sampling density is oriented in the direction perpendicular to the direction.

10. The method according to any one of claims 1 to 9, wherein, The k-space blade with the minimum k-space sampling density is oriented in a direction in which the patient's body extends furthest from the center of the examination volume of the MR system (1).

11. The method according to any one of claims 1 to 10, wherein, Low-resolution MR images are reconstructed for each k-space leaf (21-26) based on the MR signals acquired from the common central region (27) of the k-space, wherein the low-resolution MR images are compared with each other to estimate and compensate for patient movement.

12. A magnetic resonance (MR) system configured to perform the method according to any one of claims 1 to 11, wherein the MR system (1) comprises: At least one main magnet coil (2) is used to generate a uniform and stable magnetic field B0 within the inspection volume; Multiple gradient coils (4, 5, 6) for generating switched magnetic field gradients in different spatial directions within the examination volume; at least one RF coil (9) for generating RF pulses within the examination volume; multiple RF receiving antennas (11, 12, 13) with different spatial sensitivity distributions for receiving MR signals from the body (10) of a patient positioned within the examination volume; a control unit (15) for controlling the timing alternation of RF pulses and switched magnetic field gradients; and a reconstruction unit (17) for reconstructing MR images based on the received MR signals.

13. A computer program comprising instructions that, when executed by a control unit (15) of a magnetic resonance (MR) system (1), cause the MR system (1) to perform the following steps: Execute an MR imaging sequence that includes multiple radio frequency (RF) pulses and switched magnetic field gradients; MR signals are acquired in parallel via multiple RF receiving antennas (11, 12, 13), among which, The MR signals are acquired alternately over time from multiple k-space blades (21-26) according to the PROPELLER scheme, wherein the k-space blades (21-26) rotate around the center of the k-space, wherein the total acquired MR signal set spans a circle (28) in the k-space, and the common central region (27) of the k-space is covered by all the k-space blades, and the acquisition of MR signals from the respective k-space blades (21-26) involves subsampling of the k-space, wherein the acceleration factor associated with the subsampling varies according to the rotation angle.

14. A computer program comprising instructions that, when executed by a computer, particularly by a reconstruction unit (15) of a magnetic resonance MR system (1), cause the computer to calculate an MR image based on MR signals acquired in parallel via a plurality of radio frequency (RF) receiving antennas (11, 12, 13), the plurality of RF receiving antennas having different spatial sensitivity distributions, wherein, The MR signals are acquired alternately over time from multiple k-space blades (21-26) according to the PROPELLER scheme, wherein the k-space blades (21-26) rotate around the center of the k-space, wherein the total acquired MR signal set spans a circle (28) in the k-space, and the common central region (27) of the k-space is covered by all the k-space blades, the acquisition of MR signals from the respective k-space blades (21-26) involves subsampling of the k-space, wherein the acceleration factor associated with the subsampling varies with the rotation angle, and wherein the MR signals acquired via the different RF receiving antennas are combined using the spatial sensitivity distribution of different RF receiving antennas (11, 12, 13) to perform parallel image reconstruction.

Images