MR IMAGING USING MULTI-ECHO K-SPACE ACQUISITION

Abstract

The invention relates to a method of MR imaging of at least an object (10) placed in an examination volume of a MR device (1). It is an object of the invention to enable fast MR imaging using a multi-echo imaging technique which is robust with respect to motion. The method of the invention comprises the steps of:generating echo signals by subjecting the object (10) to an imaging sequence,acquiring the echo signals, each echo signal being attributed to a k-space line, wherein a number of k-space lines, which are adjacently arranged in a part of k-space, are repeatedly sampled, with said number of k-space lines being sampled in a different sequential order per repetition, andreconstructing a MR image from the acquired echo signals. Moreover, the invention relates to a MR device for carrying out this method as well as to a computer program to be run on a MR device.

Claims

2. The method of claim 1, wherein signal averaging is performed by the repeated sampling of said number of k-space lines, with k-space being divided into a central k-space part and one or more peripheral k-space parts, wherein the number of signal averages in the central k-space part is higher than the number of signal averages in the peripheral k-space parts and the central k-space part is sampled with a higher k-space density than the peripheral k-space parts.

3. The method of claim 1, wherein at least one part of k-space is sampled with variable density, and the sampling density distribution of said part being different between the repetitions.

4. The method of claim 1, wherein at least some of the repeatedly sampled k-space lines are sampled in a random sequential order.

5. The method of claim 1, wherein the multi-echo imaging sequence is a spin echo sequence, of which one shot comprises: at least one RF pulse for excitation of magnetic resonance, a plurality of refocusing RF pulses, a plurality of switched magnetic field gradients for frequency-encoding of each echo signal, and a plurality of switched magnetic field gradients for phase-encoding of each echo signal.

6. The method of claim 1, wherein the multi-echo imaging sequence is a gradient echo sequence, of which one shot comprises: at least one RF pulse for excitation of magnetic resonance, a plurality of switched magnetic field gradients for refocusing magnetic resonance and for frequency-encoding of each echo signal, and a plurality of switched magnetic field gradients for phase-encoding of each echo signal.

7. The method of claim 1, wherein at least a part of k-space is incompletely sampled.

8. The method of claim 1, wherein motion information is derived from the echo signals acquired from the repeatedly sampled k-space lines.

9. The method of claim 1, wherein the MR image is reconstructed using compressed sensing.

10. The method of claim 1, wherein the imaging sequence is a Dixon sequence, in which different echo times are selected in respective repetitions wherein signal contributions from water and fat to the acquired echo signals are separated.

11. The method of claim 10, wherein the echo signals are acquired at two or more different echo times, with the echo signals being acquired at each echo time with a different sequential order of the k space_lines.

12. The method of claim 10, wherein the signal contributions from water and fat are combined in the step of reconstructing the MR image.

13. The method of claim 1, wherein the imaging sequence is an inversion recovery sequence comprising an inversion RF pulse followed by an excitation RF pulse after an inversion recovery time.

14. A magnetic resonance (MR) device including at least one main magnet coil for generating a uniform, static magnetic field within an examination volume, a number of gradient coils for generating switched magnetic field gradients in different spatial directions within the examination volume, at least one RF coil for generating RF pulses within the examination volume and/or for receiving MR signals from an object positioned in the examination volume, a control unit for controlling the temporal succession of RF pulses and switched magnetic field gradients, and a reconstruction unit, wherein the MR device is arranged to perform the method of claim 1.

15. A computer program comprising instructions stored on a non-transitory computer readable medium, wherein the computer program when executed on a magnetic resonance (MR) device, performs the method of claim 1.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The enclosed drawings disclose preferred embodiments of the present invention. It should be understood, however, that the drawings are designed for the purpose of illustration only and not as a definition of the limits of the invention. In the drawings:

[0045] FIG. 1 shows a MR device for carrying out the method of the invention;

[0046] FIG. 2 shows a diagram of k-space schematically illustrating a first embodiment of the sampling scheme of the invention;

[0047] FIG. 3 shows a diagram of k-space schematically illustrating a second embodiment of the sampling scheme of the invention;

[0048] FIG. 4 shows a diagram of k-space schematically illustrating a third embodiment of the sampling scheme of the invention;

[0049] FIG. 5 shows a diagram of k-space schematically illustrating a fourth embodiment of the sampling scheme of the invention;

[0050] FIG. 6 shows a diagram of k-space schematically illustrating a fourth embodiment of the sampling scheme of the invention;

[0051] FIG. 7 shows schematically the in-phase, out-of-phase, water and fat images with motion ghost contributions;

[0052] FIG. 8 shows a diagram of k-space schematically illustrating sampling of in-phase and out-of-phase in a Dixon measurement.

DETAILED DESCRIPTION OF THE EMBODIMENTS

[0053] With reference to FIG. 1, a MR device 1 is shown. The device comprises superconducting or resistive main magnet coils 2 such that a substantially uniform, temporally constant main magnetic field is created along a z-axis through an examination volume.

[0054] A magnetic resonance generation and manipulation system applies a series of RF pulses and switched magnetic field gradients to invert or excite nuclear magnetic spins, induce magnetic resonance, refocus magnetic resonance, manipulate magnetic resonance, spatially and otherwise encode the magnetic resonance, saturate spins, and the like to perform

[0055] MR imaging.

[0056] More specifically, a gradient pulse amplifier 3 applies current pulses to selected ones of whole-body gradient coils 4, 5 and 6 along x, y and z-axes of the examination volume. A digital RF frequency transmitter 7 transmits RF pulses or pulse packets, via a send-/receive switch 8, to a whole-body volume RF coil 9 to transmit RF pulses into the examination volume. A typical MR imaging sequence is composed of a packet of RF pulse segments of short duration which taken together with each other and any applied magnetic field gradients achieve a selected manipulation of nuclear magnetic resonance. The RF pulses are used to saturate, excite resonance, invert magnetization, refocus resonance, or manipulate resonance and select a portion of a body 10 positioned in the examination volume. The MR signals are also picked up by the whole-body volume RF coil 9.

[0057] For generation of MR images of limited regions of the body 10, a set of local array RF coils 11, 12, 13 are placed contiguous to the region selected for imaging. The array coils 11, 12, 13 can be used to receive MR signals induced by body-coil RF transmissions.

[0058] The resultant MR signals are picked up by the whole body volume RF coil 9 and/or by the array RF coils 11, 12, 13 and demodulated by a receiver 14 preferably including a preamplifier (not shown). The receiver 14 is connected to the RF coils 9, 11, 12 and 13 via send-/receive switch 8.

[0059] A host computer 15 controls the gradient pulse amplifier 3 and the transmitter 7 to generate any of a plurality of MR imaging sequences, such as echo planar imaging (EPI), echo volume imaging, gradient and spin echo imaging, fast spin echo (TSE) imaging, and the like. For the selected sequence, the receiver 14 receives a single or a plurality of MR data lines in rapid succession following each RF excitation pulse. A data acquisition system 16 performs analog-to-digital conversion of the received signals and converts each MR data line to a digital format suitable for further processing. In modern MR devices the data acquisition system 16 is a separate computer which is specialized in acquisition of raw image data.

[0060] Ultimately, the digital raw image data is reconstructed into an image representation by a reconstruction processor 17 which applies a Fourier transform or other appropriate reconstruction algorithms. The MR image may represent a planar slice through the patient, an array of parallel planar slices, a three-dimensional volume, or the like. The image is then stored in an image memory where it may be accessed for converting slices, projections, or other portions of the image representation into appropriate format for visualization, for example via a video monitor 18 which provides a man-readable display of the resultant MR image.

[0061] With continuing reference to FIG. 1 and with further reference to FIGS. 2-5, embodiments of the motion-robust partial averaging approach of the invention are explained.

[0062] According to the invention, the body 10 is subjected to a number of shots of a multi-echo imaging sequence, such as, for example, a TSE sequence. A train of echo signals is generated by each shot of the sequence. The echo signals are acquired, wherein each echo signal is attributed to a k-space line. The k-space lines are represented in FIGS. 2-5 as thin horizontal lines extending parallel to the k.sub.x axis. Signal averaging is performed by repeatedly sampling at least some of the k-space lines. Repeated sampling means that identical k-space lines are sampled two or more times during the signal acquisition process for reconstructing the final MR image, wherein a signal to noise improvement is achieved by taking the average of the several echo signals acquired from the same k-space positions respectively. The number of averages is also referred to as the number of acquisitions (NSA). The approximate amount of improvement in signal to noise (SNR) is calculated as the square root of the number of averages.

[0063] As shown in FIG. 2, k-space 20 is divided into a central k-space part 21, a first peripheral k-space part 22, and a second peripheral k-space part 23. In the depicted embodiment, the central k-space part 21 is sampled by each shot of the TSE sequence, first peripheral k-space part 22 is sampled by every 2.sup.nd shot, while second peripheral k-space part 23 is sampled by every 4.sup.th shot. According to the proposed partial averaging approach, a variable repeated sampling of k-space positions is performed, wherein central k-space 21 is acquired more often compared to the peripheral parts 22, 23. The central k-space part 21 with its limited set of k-space lines is sampled very frequently. Motion information may be derived from the echo signals acquired from the repeatedly sampled k-space lines of central part 21. For example, cross correlations of the k-space data acquired during different averaging steps may be computed. On this basis, an automated motion compensation may be performed. In case the detected motion is found to be too severe, the corresponding echo signal data may be rejected. In case the position of the imaged object is found to have irreversibly changed due to motion, the scan parameters may be updated automatically according to the position shift. The intermediate peripheral k-space part 22 with a larger set of k-space lines is scanned less frequently to average out motion, compensate for artefacts and signal decay effects like T.sub.2 relaxation, and to provide at least a certain improvement of SNR. The second peripheral part 23 is acquired only once. It may be acquired even less than once depending on whether, for example CS and/or partial matrix techniques are used. Finally, a MR image is reconstructed from the acquired echo signals. This general approach can be applied to all types of sequences, like FFE, SE, TFE, TSE and EPI sequences.

[0064] A further aspect of the invention, which is illustrated in FIGS. 3-5, is that k-space sampling is performed with variable density, with a denser sampling of k-space in the central k-space part 21 and a sparser sampling of the peripheral k-space parts 22, 23. As illustrated in FIGS. 3-5, the acquired k-space lines are irregularly distributed in the phase-encoding direction k.sub.y. The variable k-space density effects a reduction of ghosting by avoiding a regular distribution of motion-corrupted k-space samples in k-space 20. This variable density acquisition is especially advantageous in turbo sequences like TSE or TFE, wherein the image contrast is preserved and is essentially independent of the choice of the number of acquisitions in the peripheral k-space parts 22, 23. As in the embodiment shown in FIG. 2, the central k-space part is updated more often, whereas the peripheral k-space parts 22, 23 are updated with a decreased frequency. Consequently, the most relevant, signal dominating k-space lines from central k-space 21 are updated within a short time period (for example within each shot of the multi-echo imaging sequence). In addition, the size of the k-space parts 21, 22, 23 increases to outer k-space, which results in a further distribution of motion induced k-space discontinuities over k-space 20.

[0065] FIG. 4 illustrates the k-space lines acquired during a first and second acquisition step designated as ACQ1 and ACQ2 respectively. In this embodiment, the positions of the k-space lines in the peripheral k-space part 22 are shifted in the second acquisition step ACQ2 in the k.sub.y-direction. The k.sub.y positions of the k-space lines in the first acquisition step ACQ1 are indicated as dotted lines in the right diagram of FIG. 4 for the purpose of illustration. In the central k-space part 21, in contrast, the k.sub.y positions of the k-space lines scanned during the first and second acquisition steps ACQ1 and ACQ2 are maintained, resulting in a correspondingly repeated sampling of the k-space lines in the central k-space part 21, while each k-space line in the peripheral k-space part 22 is scanned only once. During each individual acquisition step ACQ1, ACQ2, the peripheral k-space part 22 is incompletely sampled, which means that the sampling density in the k.sub.y direction is below the Nyquist threshold. The combination of the k-space lines sampled in both acquisition steps ACQ1, ACQ2 results in a completely sampled data set.

[0066] The main aspect of the invention is the repeated sampling of the respective k-space lines in a different order in each repetition, wherein the sampling order defines the temporal order in which the k-space lines are sampled. Preferably, a different random order is applied per acquisition step. The technique of the invention, which is applied in combination with all embodiments of the invention illustrated in FIGS. 2-5, averages out motion artefacts and distributes motion-induced discontinuities of the acquired k-space data resulting in an irregular k-space spacing of motion-corrupted k-space data. Ghosting in the reconstructed MR image is avoided in this way. For example, with reference to FIG. 4, the (preferably random) temporal order in which the k-space lines are sampled from the central k-space part 21 during acquisition step ACQ2 is different from the temporal sampling order applied in the central k-space part 21 during acquisition step ACQ1.

[0067] In the embodiment illustrated in FIG. 5, a random distribution of the k-space sampling grid is applied in the peripheral k-space part 22, wherein those k-space lines are sampled during acquisition step ACQ2 that have been missed out during acquisition step ACQ1. The combination of both acquisitions ACQ1, ACQ2 results in a fully sampled data set. Like in the embodiment of FIG. 4, central k-space part 21 is fully sampled during each acquisition step ACQ1, ACQ2 such that repeated sampling is achieved. The random undersampling of the peripheral k-space part 22 during each acquisition step ACQ1, ACQ2 results in a yet further improved distribution of motion-corrupted k-space samples over k-space and, consequently, in a further improved motion robustness.

[0068] In the embodiment shown in FIG. 6, again k-space 20 is divided into a central k-space part 21 and a peripheral k-space part 22. Central k-space part 21 is repeatedly sampled by correspondingly repeated application of a single-echo spin echo (SE) sequence in the depicted embodiment. The central k-space part 21 comprises a limited number of adjacently arranged k-space lines that are designated as K1, K2, K3, K4, K5. Each k-space line K1, K2, K3, K4, K5 corresponds to an echo signal generated by the SE sequence. The k-space lines K1, K2, K3, K4, K5 are repeatedly sampled with a different sequential order per repetition according to the invention. The sequential order may be varied as follows during, for example, four repetitions:

[0069] Repetition no. 1: K1, K2, K3, K4, K5;

[0070] Repetition no. 2: K2, K1, K4, K3, K5;

[0071] Repetition no. 3: K2, K4, K1, K5, K3;

[0072] Repetition no. 4: K3, K5, K2, K4, K1.

[0073] Signal averaging is performed on this basis to average out motion, compensate for artefacts and signal decay effects like T.sub.2 relaxation, and to provide an improvement of SNR. The peripheral k-space part 22 is scanned only once. Finally, a MR image is reconstructed from the acquired echo signals.

[0074] In Dixon-type measurements, motion ghosts are typically encoded in the water or fat image dependent on the spectral source as shown for the simple in-phase out-phase Dixon example of FIG. 7, (W=IP+OP, F=IPOP). Here water motion leads to a motion ghost) with the full amplitude in the water image and not in the fat image. The left portion of FIG. 7 relates to uncorrelated motion, the right portion of Figure relates to correlated motion.

[0075] In case motion is uncorrelated as shown for the simple in-phase out-phase Dixon example below left the motion ghost is present in both water and fat images, nevertheless the artefact is reduced by a factor of 2. This averaging effect can be used to effectively reduce motion artefacts.

[0076] In DIXON TSE the different profile ordering per Dixon acquisition can be achieved by a different, for example, randomized ordering of shots as illustrated below. Per shot a profile is acquired per k-space segment (in the example of FIG. 8) 7 profiles out of the 7 k-space segments are sampled). The unequidistant or random sampling of profiles (blue lines) in a shot is varied for the Dixon acquisitions e. g. for IP and OP acquisitions. In case of motion this different sampling per Dixon acquisition leads to a motion ghost that is uncorrelated in the IP and OP acquisition. Applying Dixon reconstruction this motion ghost is averaged out.

[0077] According to one aspect of the invention k-space parts are randomly ordered with a different sequential, preferably random, order per Dixon acquisition.

[0078] For turbo sequences, like TSE or TFE, profiles acquired over shots and hence positioned close in k-space within k-space segments are randomly distributed and this random order changes with the Dixon acquisition.

[0079] For non-turbo sequences like FFE, k-space is subdivided in k-space segments with a random subsegment ordering. The random subsegment ordering is changed per Dixon acquisition.

[0080] The averaging effect is enhanced the more Dixon acquisitions are used. Hence in case of Dixon quantification where typically six (6)echoes are used the averaging effect is optimized. cm 1. A method of magnetic resonance (MR) imaging of an object placed in an examination volume of a MR device, the method comprising the steps of: [0081] generating echo signals by subjecting the object to a multi-echo imaging sequence, wherein the object is subjected to a number of shots of the multi-echo imaging sequence, a train of echo signals being generated by each shot [0082] acquiring the echo signals, each echo signal being attributed to a k-space-line, wherein a number of k-space-lines, which are adjacently arranged in a part of (k.sub.x,k.sub.y)-space, is repeatedly sampled, with said number of k-space lines being sampled in a different sequential order and at different selected spin conditions per repetition, and [0083] reconstructing a MR image from the acquired echo signals.