High efficiency self-re-focusing zero echo time MR imaging

Abstract

The invention relates to a method of MR imaging of an object positioned in an examination volume of an MR device (1). It is an object of the invention to enable efficient silent ZTE imaging by self-refocusing. The method of the invention comprises the following steps: specifying a set of radial k-space spokes to cover a spherical k-space volume; selecting a predetermined number of subsets of spokes from the specified set such that the concatenation of the spokes contained in each subset forms a closed trajectory in k-space, wherein the selection of the subsets comprises optimizing a cost function; subjecting the object (10) to a zero echo time imaging sequence, wherein each of the subsets of spokes is acquired as a sequence of gradient echo signals; and reconstructing an MR image from the acquired spokes. Furthermore, the invention relates to an MR device and to a computer program for an MR device.

Description

TECHNICAL FIELD

[0001] The present invention relates to the field of magnetic resonance (MR) imaging. It relates to a MR imaging method comprising three-dimensional radial sampling in k-space. The invention further relates to a MR device and to a computer program to be run on a MR device. BACKGROUND

[0002] Image-forming MR methods are widely used nowadays which utilize the interaction between magnetic fields and nuclear spins in order to form two-dimensional or three-dimensional images, particularly in the field of medical diagnostics, as they are superior to other methods in many respects with regard to imaging soft tissue, do not require ionizing radiation and are usually non-invasive.

[0003] According to the general MR method, the body of the patient to be examined is arranged in a strong, uniform magnetic field (B0 field) whose direction simultaneously defines the axis (usually the z axis) of the coordinate system on which the measurement is based. The magnetic field produces different energy levels for individual nuclear magnetic spins in dependence on the magnetic field strength, which can be excited by means of an applied electromagnetic alternating field (RF field, also called B1 field) having a defined frequency (Larmor frequency or MR frequency). From a macroscopic point of view, the distribution of the individual nuclear magnetic spins produces a net magnetization which can be deflected out of the equilibrium state by means of suitable RF pulses, while magnetic field gradients extend in perpendicular direction to the z axis, causing the spins to perform precessional motions about the z axis. The precessional motions describe a conical surface whose aperture angle is referred to as flip angle. The magnitude of the flip angle depends on the strength and the duration of the applied electromagnetic pulse. In the case of a 90° flip angle, the nuclear magnetic spins are deflected from the z axis into the transverse plane.

[0004] After the RF pulse, the magnetization relaxes back to the initial equilibrium state, wherein the magnetization in z direction is re-established with a first time constant T1 (spin lattice relaxation or longitudinal relaxation time), and the magnetization perpendicular to the z direction relaxes with a second time constant T2 (spin-spin or transverse relaxation time). The changes in magnetization can be detected by means of one or more receive RF coils which are arranged and oriented within the examination volume of the MR device in such a manner that the changes in magnetization are measured in a direction perpendicular to the z axis. The decay of the transverse magnetization is accompanied by, for example, a transition of the nuclear magnetic spins (caused by magnetic field inhomogeneities) from an ordered state with the same phase after application of a RF pulse with 90° flip angle to a state in which all phase angles are uniformly distributed (dephasing). The dephasing can be compensated by means of a refocusing RF pulse, for example a RF pulse with 180° flip angle. This results in an echo signal (spin echo) in the receive coils.

[0005] For spatially resolving in the patient's body, linear magnetic field gradients extending along the main axis are superposed to the uniform magnetic field, causing a linear spatial dependence of the nuclear magnetic resonance frequency. The signal picked up in the receiving antenna then comprises components of different frequencies, which can be associated with different locations in the patient's body / object. The MR signals received by the radio frequency coil correspond to the spatial frequency domain, called k-space. The data are usually collected in k-space along a multitude of lines acquired with different phases. Each line is digitized by collecting a number of samples. The set of k-space data is converted into an MR image by means of a Fourier transformation or other suitable image reconstruction algorithm.

[0006] MR imaging of tissues with very short transverse relaxation times, such as bone or lung, is becoming increasingly important. Almost all known methods for this purpose essentially employ three-dimensional (3D) radial k-space sampling. In so-called zero echo time (ZTE) techniques, the readout magnetic field gradient is set before the magnetic resonance is excited, with a high bandwidth and a short and hard RF pulse. In this way, the frequency encoding starts immediately after the excitation of the magnetic resonance. The acquisition of the free induction decay (FID) signal along a radial k-space spoke starts immediately after the RF pulse, resulting in an effective echo time (TE) of zero. After the acquisition, only little time is needed for setting the next readout magnetic field gradient before the next RF pulse can be applied, enabling a very short repetition time (TR). The direction of the readout magnetic field gradient is gradually changed from repetition to repetition until a spherical volume in k-space is sampled to the desired extent, where each acquired spoke starts from the center of k-space. Without the need to switch off the readout magnetic field gradient between TRs, ZTE imaging can be performed almost silently.

[0007] Recently, self-refocusing ZTE imaging has been proposed (see, e.g., US 2017 / 0307703 Al). In self-refocusing ZTE imaging, gradient echo refocusing is added to conventional ZTE imaging. As in conventional ZTE imaging, the amplitude of the readout magnetic field gradient remains constant and only the direction is updated between repetitions until a complete spherical volume in k-space is sampled, resulting in silent three-dimensional radial imaging. RF excitation can also be achieved by a short RF pulse in the presence of the readout magnetic field gradient. The self-refocusing ZTE imaging sequence is then organized into a number of (two or more) segments, and each segment is divided into a number of cycles. Each cycle acquires a number of radial k-space spokes. Radio frequency excitation is performed only in the first cycle (FID acquisition cycle) and not in the subsequent second and further cycles (gradient echo acquisition cycles). The radial k-space spokes of each cycle cumulatively form a closed trajectory in k-space. In this way, the second and further cycles form gradient echoes of the initial FID excited in the first cycle. The TE of these gradient echoes is given by

[0008] TE = (e - 1) * N * TR,

[0009] where e is the loop counter and N is the number of radial k-space spokes acquired in each loop. TR is usually very short (on the order of 1 ms). While the magnitude of the readout magnetic field gradient remains constant throughout the acquisition, the direction updates applied between repetitions are typically larger than in conventional ZTE imaging. However, in order for the self-refocusing ZTE imaging system to remain silent, these directional updates should be as small as possible in one.

[0010] In order to sample the desired spherical k-space volume, known self-refocusing ZTE techniques typically rely on separating the rotation of radial k-space spokes into two rotations around orthogonal k-space axes. First, a plurality of segments (also referred to as interleaves) are acquired from a planar circular region in k-space, which represents a cross-section of the complete spherical k-space volume. Within the circular region, the connected spokes form a closed (polygonal) k-space trajectory to enable the acquisition of gradient echoes. The circular region is then rotated around an axis in the plane of the circular region to acquire further segments from the rotated circular region. This is continued until the complete spherical k-space volume is covered. In this way, the number of spokes required to acquire the complete spherical k-space volume is large, however, and the spokes are distributed unevenly, with unnecessary oversampling in some regions. SUMMARY

[0011] From the above it is readily appreciated that there is a need for an improved method of ZTE imaging. It is an object of the present invention to enable efficient silent ZTE imaging by self-refocusing.

[0012] According to the present invention, a method of MR imaging of an object positioned in an examination volume of an MR device is disclosed. The method of the present invention comprises the following steps:

[0013] specifying a set of radial k-space spokes to cover a spherical k-space volume;

[0014] selecting a predetermined number of subsets of spokes from the specified set such that the connected spokes contained in each subset form a closed trajectory in k-space, wherein the selection of the subsets comprises optimizing a cost function;

[0015] subjecting the object to a zero echo time imaging sequence, wherein each of the subsets of spokes is acquired as a sequence of gradient echo signals; and

[0016] reconstructing an MR image from the acquired spokes.

[0017] According to the present application, first a set of radial k-space spokes is specified to cover a spherical k-space volume (step a). The spherical k-space volume is determined by the desired field of view (FOV). As mentioned above, the radial k-space spokes start at the k-space origin and end at the surface of the sphere. The minimum density of the spoke end points on the sphere is determined by the desired (acquired) spatial resolution of the MR image to be reconstructed. Many possible approaches to determine the distribution of the end points can be envisaged. The initial specification of the set of radial k-space spokes has the advantage over previously known methods that the number of spokes can be close to the minimum number of spokes required for a certain FOV and spatial resolution, thereby reducing the scan time. Furthermore, the distribution of the end points can be more uniform, thereby improving the signal-to-noise ratio (SNR).

[0018] As a next step (step b), each of the subsets comprising a predetermined number of spokes is selected from the specified set of radial k-space spokes. The predetermined number corresponds to the number of spokes acquired within one segment of a ZTE imaging sequence. The selection is performed in such a way that the concatenation of the spokes contained in each of the subsets forms a closed trajectory in k-space. This is a prerequisite for generating the above-mentioned gradient echo signals. Furthermore, a cost function is taken into account in the selection procedure. The cost function preferably depends on the relative orientation of the spokes of each of the subsets. The relative orientation of the spokes essentially determines the acoustic noise generated by these spokes during acquisition. The gist of the present application is that the spokes in each of the subsets (from the specified set of radial k-space spokes) are selected such that (i) the generation of echoes occurs and (ii) the acoustic noise is minimized. In principle, any prior knowledge about the acoustic noise generated by the gradient switching on a particular MR device or a particular type of MR device can be incorporated into the cost function such that the generation of acoustic noise is penalized during the acquisition of the set of radial k-space spokes.

[0019] The actual acquisition of the set of radial k-space spokes with the ZTE imaging sequence is then performed in step c).

[0020] Finally, in step d), the MR image is reconstructed from the acquired radial k-space spokes.

[0021] In other words, the key aspect of the present application is to minimize the maximum angle between the successively acquired spokes within the same segment and to maximize the uniformity of the distribution of all spoke end points on the surface of the covered spherical k-space volume while ensuring that the concatenation of the spokes of each subset / segment forms a closed trajectory in k-space. That is, according to the present application, the selection of the spoke subsets based on the cost function will drive the selection to prefer smaller maximum angles between the successively spokes under the constraint that the concatenation of the spokes of each subset forms a closed trajectory in k-space, smaller variation of the end point distribution.

[0022] Ideally, the selection of the predetermined number of spokes for each of the subsets is based on a global optimum of the cost function for the entire set of acquisitions. However, it is often preferred to find a local optimum that is sufficiently close to the global optimum within a reasonable computation time. Greedy selection strategies known in the art can be used for this purpose. They can be applied to sequentially select the individual subsets of spokes, and / or the individual spokes or pairs of spokes for each of these subsets.

[0023] In a preferred embodiment of the invention, the selection of the subsets includes determining the sequence in which the spokes comprised in each subset are acquired. The acoustic noise caused by setting up the readout magnetic field gradient between the acquisition of individual radial k-space spokes depends on the relative orientation of the spokes acquired consecutively within the same segment. The selection of the subsets should therefore not only include the selection of the spokes to be acquired in each segment, but also their order.

[0024] In a preferred embodiment of the method of the invention, the set of radial k-space spokes includes pairs of spokes having antiparallel orientation. On this basis, the subsets can be selected to include pairs of spokes having antiparallel direction. This guarantees that the connection of the spokes comprised in each subset forms a closed trajectory in k-space and thereby generates a gradient echo.

[0025] Not only gradient echoes but also FID signals can be acquired and used for improving the reconstruction of the MR image, e.g. determining the spin density and / or spatial variation of the main magnetic field. Furthermore, they are able to reduce noise and / or T2 * weighting. Furthermore, correction for motion occurring between the acquisition of individual segments can be based on the FID signals, e.g. by using them as navigators for motion detection or motion compensation.

[0026] The method of the invention described so far can be performed with the aid of an MR device comprising 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 a body of a patient positioned within the examination volume, a control unit for controlling the temporal succession of RF pulses and switched magnetic field gradients, and a reconstruction unit. The method of the invention is preferably implemented by corresponding programming of the reconstruction unit and / or the control unit of the MR device.

[0027] The method of the application can advantageously be implemented by most MR devices currently used in clinical practice. For this purpose, it is only necessary to use a computer program which controls the MR device such that it carries out the above-explained method of the application. The computer program can be present on a data carrier or can be present on a data network, so that it can be downloaded for installation in the control unit of the MR device. BRIEF DESCRIPTION OF DRAWINGS

[0028] The accompanying drawings disclose preferred embodiments of the application. It is to be understood, however, that the drawings are designed solely for purposes of illustration and

[0029] Figure 1 An MR device for carrying out the method of the application is schematically shown;

[0030] Figure 2 A set of radial k-space spokes according to the application is illustrated;

[0031] Figure 3 A diagram indicating the relative orientation between successively acquired spokes according to the application is shown;

[0032] Figure 4 A further diagram indicating the relative orientation between successively acquired spokes according to another embodiment of the application is shown. DETAILED DESCRIPTION

[0033] Reference is made to Figure 1 An MR device 1 which can be used to carry out the method of the application is shown in Fig. 1. The device comprises a superconducting or resistive main magnet coil 2 which creates a substantially uniform, spatially constant main magnetic field Bo along the z-axis through an examination volume. The device further comprises a set of (first, second, and, if applicable, third) shim coils 2' in which the current flowing through the individual shim coils of the set 2' is controllable in order to minimize Bo inhomogeneities within the examination volume.

[0034] Magnetic resonance generation and manipulation systems apply a series of radio frequency pulses and switched magnetic field gradients to excite, invert or saturate nuclear magnetic spins, to induce, refocus and manipulate magnetic resonances, to spatially or otherwise encode the magnetic resonances, etc. to perform MR imaging.

[0035] More specifically, the gradient amplifiers 3 apply current pulses to selected whole-body gradient coils 4, 5 and 6 along the x, y and z axes of the examination volume. A digital RF transmitter 7 sends RF pulses via a transmit / receive switch 8 to a body RF coil 9 to send RF pulses to the examination volume. A typical MR imaging sequence consists of short duration RF pulses which, together with any applied magnetic field gradients, effect a selected operation on the nuclear magnetic resonance, including selection of the portion of the body 10 positioned in the examination volume. The MR signals are also picked up by the body RF coil 9.

[0036] For generating MR images of a limited region of the body 10, a set of local array RF coils 11, 12, 13 is placed in close proximity to the region to be imaged by means of parallel imaging. The array coils 11, 12, 13 can be used to receive MR signals induced by RF transmission with the body RF coil.

[0037] The resulting MR signals are picked up by the body 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 the transmit / receive switch 8.

[0038] A host computer 15 controls the current flow through the shim coils 2' as well as the gradient pulse amplifiers 3 and the RF transmitter 7 to generate a ZTE imaging sequence according to the present application. The receiver 14 receives MR signals from the respective radial k-space spokes after a fast succession of RF excitation pulses. A data acquisition system 16 performs an analog-to-digital conversion on the received MR signals and converts them into a digital format suitable for further processing. In modern MR devices, the data acquisition system 16 is a separate computer which is dedicated to the acquisition of raw image data.

[0039] Ultimately, the digital raw image data is reconstructed into an image representation by a reconstruction processor 17 which applies appropriate reconstruction algorithms. The MR image represents a three-dimensional volume. The image is then stored in an image memory where it can be accessed for converting projections or other portions of the image representation into a suitable format for visualization, e.g. via a video monitor 18 which provides a human readable display of the resulting MR image.

[0040] The essence of the nearly silent ZTE technique applied by the present invention is to transmit RF excitation pulses while the frequency-encoded readout magnetic field gradient is turned on. The readout magnetic field gradient is not intended to be a slice-selective gradient, which means that the RF excitation pulses must be very short (typically on the order of 1 μs or 10 μs) to achieve sufficient excitation bandwidth. Alternatively, RF excitation pulses with frequency scanning can be applied. In the presence of the readout magnetic field gradient, the readout of the FID signal occurs during the interval immediately following the RF excitation pulse. These intervals are also preferably short (typically on the order of 100 μs or 1 ms). The intensity and direction of the readout magnetic field gradient remain essentially constant within each excitation / readout cycle to acquire MR signals from one radial k-space spoke. After each excitation / readout cycle, the direction is changed. For complete sampling of k-space, this operation is repeated until the spherical volume is completely covered with sufficient density.

[0041] According to the present invention, self-refocusing ZTE imaging is achieved through a gradient echo refocusing mechanism. The pulse sequence is organized into several (two or more) segments, and each segment is divided into several cycles. Each cycle packet acquires several radial k-space spokes. RF excitation is only effective for the first cycle (FID acquisition cycle) and is then turned off in the subsequent second and another cycle (gradient echo acquisition cycle). The radial k-space spokes of each cycle form a closed trajectory in k-space. In this way, the subsequent cycles form the gradient echoes of the initial FID excited in the initial cycle. For details of the self-refocusing ZTE imaging sequence adopted by the present invention, see US2017 / 0307703A1.

[0042] The present invention proposes to first specify a set of radial k-space spokes to cover the spherical k-space volume to be acquired. In a possible embodiment, the set of spokes is specified as follows:

[0043] Define a sufficient number N on the surface of the k-space sphere P parallel lines, such as

[0044]

[0045] where N corresponds to the diameter of the k-space sphere, discretized according to the acquired spatial resolution.

[0046] Sufficient number of endpoints N E Place them equidistantly along each parallel line, e.g.

[0047]

[0048] By targeting N P Choose odd values ​​and for N EThe even values are chosen, ensuring the existence of anti-parallel radial spoke pairs. Such predefined sets of endpoints of radial k-space spokes are plotted in the illustration of Figure 2 Fig. 1.

[0049] In the next step, pairs of anti-parallel spokes are selected from the specified set under optimization. This way, the k-space trajectory resulting from the radial k-space spokes of one subset corresponding to one acquisition segment is ensured to be self-refocusing.

[0050] In the optimization, the following is repeated until all radial k-space spokes are assigned to a subset:

[0051] A previously unselected spoke is picked from the specified set either systematically or randomly.

[0052] The cost function is evaluated for all permutations of this spoke and N S / 2-1 unselected spokes, where N S denotes the number of spokes in each subset.

[0053] A greedy selection strategy is applied, selecting the best permutation (for which the cost function takes its minimum value, for example) and assigning the respective spokes and their anti-parallel counterparts to the current subset.

[0054] As a cost function, for example, the square of the relative orientation angle between consecutive spokes can be employed in order to achieve a minimum level of acoustic noise during ZTE acquisition.

[0055] The greedy selection strategy, which is known per se, aims at a locally optimal selection for each subset with the goal to find a global optimum for the entire specified set. The greedy optimization strategy can in fact not find the true global optimum solution, but nevertheless it still generates a locally optimal solution for the subsets, which at least approximates the global optimum solution within a reasonable computation time. The global optimum solution globally minimizes the acoustic noise level not only for each subset or acquisition segment, but for the entire specified set or the entire acquisition.

[0056] However, instead of the greedy selection strategy, any known heuristic algorithm can be applied for an efficient but approximate optimal permutation search. In particular, the search can be stopped as soon as a predetermined maximum value of the relative orientation angle between consecutive spokes within a segment is met.

[0057] Figure 3Representative results obtained with the above procedure are provided. In this example, each subset selected eight spokes, with an optimal angle of 45° between successive spokes. Using a greedy selection strategy, the actual angle between successive spokes was close to the optimal value for most spokes. Only towards the end of the procedure, the selection of the remaining unselected spokes became very limited, and the angle between successive spokes significantly exceeded the optimal value.

[0058] If the predetermined maximum value of the relative orientation angle between successive spokes in a subset cannot be met anymore, spokes can be added at the cost of a small increase in the number of spokes required and the corresponding scan time. This is demonstrated in Figure 4 where reacquiring spokes is allowed, i.e. some spokes are assigned to a subset that has already been selected for a previous subset. If the relative orientation angle exceeds 50°, spokes are allowed to be reselected, resulting in an increase of 13% in the total number of spokes.

Claims

1. A method for MR imaging of an object (10) positioned in an examination volume of an MR device (1), the method comprising the following steps: Specifying a set of radial k-space spokes to cover a spherical k-space volume; selecting subsets, wherein each subset includes a predetermined number of spokes from the specified set such that a connection of the spokes contained in each of the subsets forms a closed trajectory in k-space, wherein the selection of the subsets is based on an optimization cost function that prefers (i) minimizing a maximum angle between successively acquired spokes within the same subset and (ii) maximizing a uniformity of distribution of endpoints of all spokes on a surface of a covered spherical k-space volume while ensuring that a connection of the spokes in each subset forms a closed trajectory in k-space; subjecting the subject (10) to a zero echo time imaging sequence, wherein each of the subsets of spokes is acquired by executing a gradient echo acquisition cycle and a free induction decay acquisition cycle for acquiring a free induction decay signal; and An MR image is reconstructed from the acquired spokes.

2. The method according to claim 1, wherein The selection of the subsets comprises determining a sequence in which the spokes contained in each of the subsets are acquired.

3. The method according to claim 2, wherein: The cost function depends on the relative orientations of successively collected spokes from the same subset.

4. The method according to any one of claims 1 to 3, wherein: The cost function penalizes the acoustic noise level generated during acquisition of the subset of spokes.

5. The method according to any one of claims 1 to 4, wherein: The set of radial k-space spokes comprises pairs of spokes having anti-parallel orientations.

6. The method according to claim 5, wherein: The subset is selected to include pairs of spokes having antiparallel orientations.

7. The method according to any one of claims 1 to 6, wherein: A greedy selection strategy is used in the sequential selection of a subset of spoke individuals.

8. The method according to any one of claims 1 to 7, wherein: A greedy selection strategy is used to sequentially select individual spokes or pairs of spokes for each of the subsets.

9. An MR device comprising: at least one main magnet coil (2) for generating a uniform static magnetic field within an examination volume; Several 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 and / or for receiving MR signals from an object (10) positioned within the examination volume; a control unit (15) for controlling the temporal succession of RF pulses and the switched magnetic field gradients; and a reconstruction unit (17), wherein the MR device (1) is arranged to perform the following steps: Specifying a set of radial k-space spokes to cover a spherical k-space volume; selecting subsets of a predetermined number of spokes from the specified set such that a connection of the spokes contained in each of the subsets forms a closed trajectory in k-space, wherein the selection of the subsets is based on a cost function based on optimizing the cost function with a preference for (i) minimizing a maximum angle between successively acquired spokes within the same subset and (ii) maximizing a uniformity of distribution of endpoints of all spokes on the surface of the covered spherical k-space volume while ensuring that a connection of the spokes in each subset forms a closed trajectory in k-space; subjecting the subject (10) to a zero echo time imaging sequence, wherein each of the subsets of spokes is acquired by executing a gradient echo acquisition cycle and a free induction decay acquisition cycle for acquiring a free induction decay signal; and An MR image is reconstructed from the acquired spokes.

10. A computer program product storing a computer program to be run on an MR device, the computer program comprising instructions for: Specifying a set of radial k-space spokes to cover a spherical k-space volume; A predetermined number of subsets of spokes are selected from the specified set such that a connection of the spokes contained in each of the subsets forms a closed trajectory in k-space, wherein The selection of the subsets is based on a cost function that prefers (i) minimizing the maximum angle between successively acquired spokes within the same subset and (ii) maximizing the uniformity of the distribution of the endpoints of all spokes on the surface of the covered spherical k-space volume while ensuring that the connections of the spokes in each subset form closed trajectories in k-space; generating a zero echo time imaging sequence in which each of the subsets of spokes is acquired by performing a gradient echo acquisition cycle and a free induction decay acquisition cycle for acquiring a free induction decay signal; and An MR image is reconstructed from the acquired spokes.

Images