METHOD AND SYSTEM FOR CONTROLLING A MAGNETIC RESONANCE IMAGING SYSTEM

Abstract

A method for controlling a magnetic resonance imaging system, including: selecting a plurality of spatially non-selective initial RF-pulses each having a predefined pulse shape and a predefined frequency; determining a combined RF-pulse from the initial RF-pulses by choosing a time-offset comprising a relative application time-shift between the initial RF-pulses, wherein this time-offset is chosen such that the initial RF-pulses overlap; and including the combined RF pulse in a pulse sequence applied in a magnetic resonance imaging system.

Claims

1. A method for controlling a magnetic resonance imaging system, comprising: selecting a plurality of spatially non-selective initial RF-pulses each having a predefined pulse shape and a predefined frequency; determining a combined RF-pulse from the initial RF-pulses by choosing a time-offset comprising a relative application time-shift and a phase-shift between the initial RF-pulses, wherein the time-offset is chosen such that the initial RF-pulses overlap; and including the combined RF pulse in a pulse sequence applied in a magnetic resonance imaging system.

2. The method according to claim 1, wherein the initial RF-pulses are designed for spatially non-selective excitation of proton spins, for pulse sequences designed for magnetization transfer, chemically selective saturation, or chemical exchange saturation transfer.

3. The method according to claim 1, wherein at least two of the initial RF-pulses have a different frequency with separate, non-overlapping frequency bands, and the difference between the frequencies of the initial RF-pulses is more than 50 Hz.

4. The method according to claim 1, wherein the time-offset is chosen such that an absolute value of a maximum of the combined RF-pulse does not exceed a predefined maximum RF-intensity being lower than the maximum applicable RF-intensity of the magnetic resonance imaging system or the absolute value of a maximum RF-intensity of the initial RF-pulses.

5. The method according to claim 1, wherein the pulse shape B1(t) of a number of initial RF-pulses with the amplitude A follows the formula $B_1\left(t\right)=A{e}^{-{\left(t-t_0\right)}^2/{\sigma }^2}{e}^{\mathrm{i\Delta \Delta \omega }}\mathrm{or}{B}_1\left(t\right)=A\sin \left(\frac{t-t_0}{\sigma }\right)\frac{\sigma }{t-t_0}{e}^{\mathrm{i\Delta \omega t}}.$

6. The method according to claim 1, wherein the pulse shape and/or the duration of a number of initial RF-pulses is identical, and these initial RF-pulses have different frequency offsets.

7. The method according to claim 1, wherein two initial RF-pulses are arranged such that there is always a temporal overlap of the initial RF-pulses having a non-empty set of time points where an RF-contribution of both initial RF-pulses is non-zero.

8. The method according to claim 1, wherein the time-offset is chosen such that a minimal temporal shift between two initial RF-pulses is determined, where the absolute value of a maximum of the combined RF-pulse does not exceed the predefined maximum RF-intensity, with the steps: a) providing the pulse shapes of the initial RF-pulses; b) providing a predefined minimal test-offset comprising a time-shift and a phase-shift with the value zero; c) providing a predefined maximum RF-intensity; d) calculating a summed RF-pulse of the initial RF-pulses, where at least one initial RF-pulse is temporally shifted with the test-offset; and e) comparing the absolute value of the maximum of the summed RF-pulse with a predefined maximum RF-intensity, and if the absolute value of the maximum of the summed RF-pulse exceeds the maximum RF-intensity, increase the test-offset between two of the initial RF-pulses with a predefined temporal value and repeat steps d) to e), and if it does not, take the actual summed RF-pulse as combined RF-pulse, wherein the steps are performed until the time-shift of the test-offset exceeds the length of the initial RF-pulses such that they no longer overlap.

9. The method according to claim 8, wherein in the course of increasing the test-offset, a time-shift is increased by a predefined positive or negative shift, or a phase-shift is increased by a positive or negative shift until it exceeds the value of 2π, wherein the phase-shift is increased in an inner loop and the time-shift is increased in an outer loop.

10. A system for controlling a magnetic resonance imaging system, comprising: a selector designed to select a plurality of spatially non-selective initial RF-pulses each having a predefined pulse shape and a predefined frequency; a determiner designed to determine a combined RF-pulse from the initial RF-pulses by choosing a time-offset comprising a relative application time-shift and a phase-shift between the initial RF-pulses, wherein this time-offset is chosen such that the initial RF-pulses overlap; and a sequence controller designed to include the combined RF pulse in a pulse sequence applied by a radio-frequency transmitter in a magnetic resonance imaging system.

11. A controller for controlling a magnetic resonance imaging system comprising a system according to claim 10.

12. A magnetic resonance imaging system comprising a controller designed to control a magnetic resonance imaging system comprising a system according to claim 10.

13. A non-transitory computer-readable medium on which is stored program elements that are readable and executed by a computer in order to perform steps of the method according to claim 1 when the program elements are executed by the computer.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0070] FIG. 1 shows a simplified MRI system according to an aspect of the disclosure.

[0071] FIG. 2 shows a block diagram of the process flow of a preferred method according to the disclosure.

[0072] FIG. 3 shows a process flow of the determination of the combined RF-pulse.

[0073] FIG. 4 shows two fluid-attenuated, T2-weighted Turbo spin echo images according to the state of the art.

[0074] FIG. 5 shows the feasibility of MT-preparations to improve white matter/grey matter contrast in brain MRI according to the state of the art.

[0075] FIG. 6 shows an example for a Gaussian RF-pulse according to the state of the art.

[0076] FIG. 7 shows an example for Gaussian RF-pulse according to the state of the art.

[0077] FIG. 8 shows an example for two adjacent Gaussian RF-pulses according to the state of the art.

[0078] FIG. 9 shows an example for a Dual-band Gaussian RF-pulse according to the state of the art.

[0079] FIG. 10 shows an example for a rectangular RF-pulse according to the state of the art.

[0080] FIG. 11 shows an example for a double Sinc RF-pulse according to the state of the art.

[0081] FIG. 12 shows an example for a Dual-band Sinc RF-pulse according to the state of the art.

[0082] FIG. 13 shows an example for multiple Gaussian RF-pulses according to the state of the art.

[0083] FIG. 14 shows an example for multiple Gaussian RF-pulses according to the state of the art.

[0084] FIG. 15 shows an example for Gaussian RF-pulses according to a preferred aspect of the disclosure.

[0085] FIG. 16 shows an example for Gaussian RF-pulses according to a preferred aspect of the disclosure.

[0086] FIG. 17 shows an example for optimized Sinc RF-pulses according to a preferred aspect of the disclosure.

[0087] FIG. 18 shows an example for multiple Gaussian RF-pulses according to a preferred aspect of the disclosure.

[0088] In the diagrams, like numbers refer to like objects throughout. Objects in the diagrams are not necessarily drawn to scale.

DETAILED DESCRIPTION

[0089] FIG. 1 shows a schematic representation of a magnetic resonance imaging system 1 (“MRI-system”). The MRI system 1 includes the actual magnetic resonance scanner (data acquisition unit) 2 with an examination space 3 or patient tunnel in which a patient or test person is positioned on a driven bed 8, in whose body the actual examination object is located.

[0090] The magnetic resonance scanner 2 is typically equipped with a main field magnet system 4, a gradient system 6 as well as an RF transmission antenna system 5 and an RF reception antenna system 7. In the shown exemplary aspect, the RF transmission antenna system 5 is a whole-body coil permanently installed in the magnetic resonance scanner 2, in contrast to which the RF reception antenna system 7 is formed as local coils (symbolized here by only a single local coil) to be arranged on the patient or test subject. In principle, however, the whole-body coil can also be used as an RF reception antenna system, and the local coils can respectively be switched into different operating modes.

[0091] The main field magnet system 4 is preferably designed that at least two slices can be recorded. It here is designed in a typical manner so that it generates a main magnetic field in the longitudinal direction of the patient, i.e. along the longitudinal axis of the magnetic resonance scanner 2 that proceeds in the z-direction. The gradient system 6 typically includes individually controllable gradient coils in order to be able to switch (activate) gradients in the x-direction, y-direction or z-direction independently of one another.

[0092] The MRI system 1 shown here is a whole-body system with a patient tunnel into which a patient can be completely introduced. However, in principle the disclosure can also be used at other MRI systems, for example with a laterally open, C-shaped housing, as well as in smaller magnetic resonance scanners in which only one body part can be positioned.

[0093] Furthermore, the MRI system 1 has a central control device 13 that is used to control the MRI system 1. This central control device 13 includes a sequence control unit 14 for measurement sequence control. With this sequence control unit 14, the series of radio-frequency pulses (RF pulses) and gradient pulses can be controlled depending on a selected pulse sequence PS or, respectively, a series of multiple pulse sequence PS to acquire magnetic resonance images of the slices within a measurement session. For example, such a series of pulse sequence PS can be predetermined within a measurement or control protocol P. Different control protocols P for different measurements or measurement sessions are typically stored in a memory 19 and can be selected by and operator (and possibly modified as necessary) and then be used to implement the measurement.

[0094] To output the individual RF pulses of a pulse sequence PS, the central control device 13 has a radio-frequency transmission device 15 that generates and amplifies the RF pulses and feeds them into the RF transmission antenna system 5 via a suitable interface (not shown in detail). To control the gradient coils of the gradient system 6, the control device 13 has a gradient system interface 16. The sequence control unit 14 communicates in a suitable manner with the radio-frequency transmission device 15 and the gradient system interface 16 to emit the pulse sequence PS.

[0095] Moreover, the control device 13 has a radio-frequency reception device 17 (likewise communicating with the sequence control unit 14 in a suitable manner) in order to acquire magnetic resonance signals (i.e. raw data) for the individual measurements, which magnetic resonance signals are received in a coordinated manner from the RF reception antenna system 7 within the scope of the pulse sequence PS.

[0096] A reconstruction unit 18 receives the acquired raw data and reconstructs magnetic resonance image data therefrom for the measurements. This reconstruction is typically performed on the basis of parameters that may be specified in the respective measurement or control protocol. For example, the image data can then be stored in a memory 19.

[0097] Operation of the central control device 13 can take place via a terminal 10 with an input unit and a display unit 9, via which the entire MRI system 1 can thus also be operated by an operator. MR images can also be displayed at the display unit 9, and measurements can be planned and started by means of the input unit (possibly in combination with the display unit 9), and in particular suitable control protocols can be selected (and possibly modified) with suitable series of pulse sequence PS as explained above.

[0098] The control device 13 comprises a system 12 designed to perform the method according to the disclosure. This system 12 comprises the following components that may appear to be software modules.

[0099] A selection unit 20 selects a number of initial RF-pulses each having a predefined pulse shape and a predefined frequency. This can be achieved when an examination is pending and a protocol P is chosen for this examination. In this exemplary protocol P there is a list of the initial RF-pulses P1, P2 that have to be applied during this examination. Thus, the protocol can easily be parsed for RF-pulses following each other and these RF-pulses can be selected as the initial RF-pulses. However, usually the combination of initial RF-pulses would take place within the MR pulse sequence already (e.g. within the “MT-preparation” module or the “fat saturation” module of the software). By taking care of the combination at an early stage, the sequence already “knows” about the reduced time of this module and can thus e.g. allow the user to take advantage of a shorter measurement duration in the protocol setup before starting the scan.

[0100] A determination unit 21 combines the selected initial RF-pulses P1, P2 according to the special method (see e.g. FIG. 2) to a combined RF-pulse CP by choosing a time-offset comprising a relative application time-shift and especially also a phase-shift between the initial RF-pulses P1, P2, wherein this time-offset is chosen such that the RF-pulses P1, P2 overlap.

[0101] The system 12 according to the disclosure is here the sum of selection unit 20, determination unit 21, sequence control unit 14 and radio-frequency transmission device 15 as shown by the dashed frame.

[0102] The MRI system 1 according to the disclosure, and in particular the control device 13, can have a number of additional components that are not shown in detail but are typically present at such systems, for example a network interface in order to connect the entire system with a network and be able to exchange raw data and/or image data or, respectively, parameter maps, but also additional data (for example patient-relevant data or control protocols).

[0103] The manner by which suitable raw data are acquired by radiation of RF pulses and the generation of gradient fields, and MR images are reconstructed from the raw data, is known to those skilled in the art and thus need not be explained in detail herein.

[0104] FIG. 2 shows a block diagram of the process flow of a preferred method according to the disclosure for controlling a magnetic resonance imaging system 1 (see. e.g. FIG. 1).

[0105] In step I, a multiplicity of initial RF-pulses P1, P2 is selected, each having a predefined pulse shape, here Gaussian, and a predefined frequency. The horizontal axis is the time and the vertical axis is the RF-intensity.

[0106] In step II, a combined RF-pulse CP is determined from the initial RF-pulses P1, P2 by choosing a time-offset comprising a relative application time-shift and especially also a phase-shift between the initial RF-pulses P1, P2 as can be seen in the box over step II. There are two initial RF-pulses P1, P2, one solid and one dashed, that are shifted in time until the maximum RF-intensity is not exceeding a predefined threshold for a maximum value M of the RF intensity. The sum SP of the two initial RF-pulses P1, P2 is shown as envelope over these two curves. The maximum value may be given by the maximum power of the scanner or chosen as the maximum of one of the initial RF-pulses P1, P2 as it is done here. At last, a time-shift and especially a phase-shift of the initial RF-pulses P1, P2 is chosen such that the RF-pulses overlap and the sum SP of the two initial RF-pulses P1, P2 does not exceed the maximum value M.

[0107] In step III, the combined RF pulse CP is included in a pulse sequence PS that is to be applied for the examination.

[0108] In step IV, the pulse sequence PS is applied in a magnetic resonance imaging system in the course of the actual examination.

[0109] FIG. 3 shows a process flow of the determination of the combined RF-pulse. It shows the following algorithm:

Δt=0; Δφ=0; define initial pulse-functions F(t) and G(t).   1.

[0110] A test-offset combined from the time-shift Δt and the phase-shift Δφ is set to zero. Furthermore, pulses F(t) and G(t), mathematical functions for the initial RF-pulses P1, P2, are defined, wherein the pulse shapes are the courses of the curves. F(t) and G(t) may have the same mathematical function (G=F) or different functions depending from the desired examination. The duration of the curves, i.e. the maximum time-shift where the curves do not overlap, is here given as T.sub.F and T.sub.G.

[0111] Furthermore, the maximum RF-amplitude RF.sub.max is e.g. given by the properties of the machine or by other boundaries. It may be the absolute value of the maximum RF-amplitude of the initial RF-pulses P1, P2, however, in this example it is higher.

F.sub.sum(t)=F(t)+G(t−Δt).Math.e.sup.iΔφ  2.

[0112] The (complex) sum function of F(t) and G(t) is calculated, where G(t) is shifted by the time-shift Δt and the phase-shift Δφ. At first, these values are zero and the two functions overlap completely, this would probably result in a maximum amplitude that exceeds RF.sub.max. Anticipating the following description, it is said that the following loops will always return to this point.

Max(|F.sub.sum(t)|)≤RF.sub.max(for all t∈{t: |F.sub.sum(t)|>0}?  3.

[0113] In this step it is determined whether the absolute value of the sum function F.sub.sum(t) exceeds the maximum value RF.sub.max. It should be noted that the curves may have positive and negative values, since RF-pulses may have positive or negative amplitudes. The determination is done over the whole duration of F.sub.sum(t), as long as this function is not zero.

[0114] In the case that Max(|F.sub.sum(t)|)≤RF.sub.max, a solution is found and the combined RF-pulse CP is F.sub.sum. The algorithm is stopped then and the combined RF-pulse CP is included into the pulse sequence PS (see above step III).

[0115] In the case, Max(|F.sub.sum(t)|) exceeds RF.sub.max, the algorithm proceeds.

Δφ=Δφ+Δφ.sub.inc.   4.

[0116] In this inner loop, the phase-shift Δφ is incriminated by a predefined incrimination step Δφ.sub.inc.

Δφ<2π?  5.

[0117] As long as Δφ<2π, the algorithm returns to step 2 (calculation of F.sub.sum with the incriminated phase-shift Δφ). Since it does not make sense to incriminate a phase-shift more than 2π, the loop ends if Δφ is equal or bigger than 2π and the algorithm proceeds with step 6.

[0118] It should be noted that a phase-shift is not always necessary or desired and only a time-shift should be performed. In this case, this inner loop is not included into the algorithm (and the term e.sup.iΔφ is omitted as well or set to 1).

Δt=Δt+Δt.sub.inc.   6.

[0119] In this outer loop, the time-shift Δt is incriminated by a predefined incrimination step Δt.sub.inc. This value may depend on the scanner hardware (e.g. the digital-analog-converters). The value Δt.sub.inc may be a certain minimum time-shift step, e.g. 100 ns or 10 ns or 1 ns.

|Δt|<(T.sub.F/2+T.sub.G/2)?  7.

[0120] As long as there is an overlap between F and G, the algorithm returns to step 2 (calculation of F.sub.sum with the incriminated time-shift Δt). Before going to step 2, the phase-shift Δφ is set to zero so that the inner loop may be performed again. When the absolute value |Δt| exceeds the condition T.sub.F/2+T.sub.G/2, there is not severe overlap any more, and the loop ends. This would result in no result (what is a very rare end, e.g. only for rectangular functions F and G with RF.sub.max=max F).

[0121] FIG. 4 shows two fluid-attenuated (FLAIR), T2-weighted Turbo spin echo (TSE) images acquired with multiple slices (left) and a single slice only (right) according to the state of the art. These pictures demonstrate the relevance of Magnetization Transfer (MT) effects, wherein all other imaging parameters were kept identical. Compared to gray matter, white matter exhibits a considerable amount of bound water protons, mostly macromolecules within the myelin sheaths of axons. Saturation of the latter by the additional RF-pulses, and magnetization transfer between the water species yield additional attenuation of the white matter signal. In this T2-FLAIR example, the disclosure would enable the more advantageous ms-EPI acquisition to achieve contrast comparable to that of the established TSE technique.

[0122] FIG. 5 shows examples demonstrating the feasibility of MT-preparations to improve the white matter to grey matter contrast in brain MRI. All images have been acquired with a fluid-attenuated (FLAIR) multi-shot echoplanar acquisition (ms-EPI), using various (standard) MT-preparation schemes. A corresponding T2-FLAIR TSE image is shown for reference.

[0123] It turns out that the amount of saturation in the bound water pool depends on the average of the squared applied B1 field amplitude <B1> (which is proportional to the applied RF power). The RF amplitude can be increased to enhance the effectiveness of the MT-preparation module, but hardware limitations restrict allowed levels of increase. Alternatively, repetitive applications of MT-preparation pulses yield enhanced MT-effects, at the expense of a longer duration.

[0124] For better understanding of the combined RF-pulses of the disclosed subject matter, some exemplary pulses of the state of the art are shown. The following FIGS. 6 to 14 show a number of different initial RF-pulses applied for various examinations. In these figures, the magnitude time course of a resulting pulse is shown in the upper left graph with arbitrary but compatible pulse magnitudes (a.u.=arbitrary units). On the lower left side, the time course of the pulse phase is shown and on the right the frequency distribution.

[0125] The single pulses have the temporal width parameter σ, the duration T (time-width at the base), a pulse energy E of E=∫.sub.t.sub.0.sub.−T/2.sup.t.sup.0.sup.+T/2B.sub.1.sup.2(t)dt with the peak RF amplitude B1 and the duration T with its center t.sub.0.

[0126] In FIGS. 6 to 9, the pulses have a Gaussian envelope (see above equation 1) as their basis, FIG. 10 a rectangular pulse, FIGS. 11 and 12 a Sinc-RF-pulse envelope (see above equation 2) and FIGS. 13 and 14 again Gaussian envelopes.

[00003]$\begin{array}{cc}\mathrm{Gaussian}\mathrm{envelop}\text{e:}{B}_1\left(t\right)=A{e}^{-{\left(t-t_0\right)}^2/{\sigma }^2}{e}^{\mathrm{i\Delta \omega t}}& \left(1\right)\\ \mathrm{Sinc}R\text{F-pulse}\mathrm{envelope}.B_1\left(t\right)=A\sin \left(\frac{t-t_0}{\sigma }\right)\frac{\sigma }{t-t_0}{e}^{\mathrm{i\Delta \omega t}}& \left(2\right)\end{array}$

[0127] FIG. 6 shows a single Gaussian pulse with a frequency offset Δω/2π of 1.2 kHz and a single-band Gaussian frequency characteristic. The disadvantage is a small MT-effect.

[0128] FIG. 7 shows an extended single Gaussian pulse with a frequency offset Δω/2π of 1.2 kHz and a single-band Gaussian frequency characteristic. The disadvantage is a long duration for only a moderate MT-effect.

[0129] FIG. 8 shows two adjacent Gaussian pulses (with frequency alternation) with frequency offsets Δω/2π of ±1.2 kHz and a dual-band Gaussian frequency characteristic. The disadvantage is a long duration for only a moderate MT-effect.

[0130] FIG. 9 shows a dual-band Gaussian pulse with frequency offsets Δω/2π of ±1.2 kHz and a dual-band Gaussian frequency characteristic. The disadvantage is a high peak RF-amplitude.

[0131] FIG. 10 shows a rectangular pulse with a frequency offset Δω/2π of +1.2 kHz and a single-band Sinc frequency characteristic. The disadvantage is besides a long duration a pre-saturation of on-resonance spins.

[0132] In the following, some properties of the examples of FIGS. 6 to 10 are listed, Disadvantageous values are marked with a “!”.

TABLE-US-00004 Figure 6 7 8 9 10 Duration T [ms]  5  10 !  10 !  5  10 Temporal Width Par. σ [ms]  1  2  1  1 Peak RF amplitude [a.u.]  10  10  10  20 !  10 Pulse energy E [a.u.] 125 ! 250 250 250 1000

[0133] With a limited RF peak amplitude, rectangular RF-pulse shapes allow to apply the maximum RF-intensity within a given duration and might thus appear as be the best choice regarding the effectiveness for generating MT-contrast. However, it turns out that the frequency spectra of rectangular RF-pulses exhibit non-negligible on-resonance contributions, which lead to undesirable pre-saturation of the hydrogen nuclei to be imaged.

[0134] The principles of FIGS. 6 to 10 may get applied to other spatially non-selective excitations resp. saturations of specific frequency bands. For example, a simultaneous chemically selective saturation of fat protons (with a chemical shift of e.g. 3 ppm resulting in a frequency offset of 380 Hz at 3 T) and of silicone protons (e.g. 5.5 ppm resulting in a frequency offset of 700 Hz at 3 T) requires the application of two RF-pulses with the desired frequency offsets and bandwidths. Here the above mentioned Sinc envelope (equation 2) are applied.

[0135] FIG. 11 shows two adjacent Sinc RF-pulses with the two frequencies 380 Hz and 700 Hz and a dual-band rectangular frequency characteristic. The disadvantage is a long duration.

[0136] FIG. 12 shows a dual-band Sinc RF-pulse with the two frequencies 380 Hz and 700 Hz and a dual-band rectangular frequency characteristic. The disadvantage is a high peak RF-amplitude.

[0137] In the following, some properties of the examples of FIGS. 11 and 12 are listed, Disadvantageous values are marked with a “!”.

TABLE-US-00005 Figure 11 12 Duration T [ms] 60 ! 30 Temporal Width Par. σ [ms]  2, 5  2, 5 Peak RF amplitude [a.u.] 10 20 !

[0138] If the actual shape of the frequency spectrum doesn't matter, as long as there is negligible pre-saturation of magnetization which shall get used for imaging purposes, it is possible to apply a series of preparation pulses, e.g. in order to amplify the magnetization transfer effect. FIGS. 13 and 14 show two state of the art examples.

[0139] FIG. 13 shows a series of 6 adjacent Gaussian RF-pulses with a frequency offset Δω/2π of 1.2 kHz and with a Single-band non-Gaussian frequency characteristic. The disadvantage is a long duration.

[0140] FIG. 14 shows a series of 6 adjacent Gaussian RF-pulses with a frequency offset Δω/2π of 1.2 kHz and with a Dual-band non-Gaussian frequency characteristic. The disadvantage is a long duration.

[0141] In the following, some properties of the examples of FIGS. 13 and 14 are listed, Disadvantageous values are marked with a “!”.

TABLE-US-00006 Figure 13 14 Duration T [ms]  30 !  30 ! Temporal Width Par. σ [ms]  1  1 Peak RF amplitude [a.u.]  10  10 Pulse energy E [a.u.] 750 750

[0142] The following FIGS. 15 to 18 show a number of combined RF-pulses CP according to preferred aspects of the disclosure. FIGS. 15 and 16 correspond to cases where FIGS. 6 to 10 are referring to, FIG. 17 corresponds to Sinc pulses (see FIGS. 11 and 12) and FIG. 18 corresponds to multiple pulses (see FIGS. 13 and 14). Again, the magnitude time course of a resulting pulse (here the combined pulse CP) is shown in the upper left graph with arbitrary but compatible pulse magnitudes. On the lower left side, the phase time course is shown and on the right the frequency distribution.

[0143] FIG. 15 shows a shifted dual-band Gaussian pulse with a dual-band Gaussian frequency characteristic. In this example, the RF-pulse comprises two sub-pulses (initial RF-pulses P1, P2, see e.g. FIG. 2) with identical Gaussian shapes and identical durations of 5 ms. The frequency offsets Δω/2π of the pulses are +1.2 kHz for one sub-pulse and −1.2 kHz for the other sub-pulse. The second sub-pulse starts 2.5 ms before the first sub-pulse ends. Due to their different frequency offsets, the overlap does not affect the frequency spectrum.

[0144] The envelope of the combined RF-pulse CP is:

[00004]$\begin{array}{cc}1.B_1\left(t\right)=A\left(e^{-{\left(t-t_0\right)}^2/{\sigma }^2}{e}^{{\mathrm{i\Delta \omega }}_{1}t}+e^{-{\left(t-t_0-\Delta t\right)}^2/{\sigma }^2}{e}^{{\mathrm{i\Delta \omega }}_{2}t}\right).& \left(3\right)\end{array}$

[0145] The duration is only 7.5 ms and the peak RF amplitude 10 a.u. with a pulse energy of 250 a.u. Thus, with the selected pulse shift, the peak amplitude of the combined RF-pulse does not exceed the single-pulse maximum.

[0146] Compared with the state of the art, the pulse energy corresponds with the pulse energies of FIGS. 7 to 9 (FIG. 6 provides not enough energy) with the advantage that the pulse duration is lower than 10 ms while the peak RF amplitude is only 10 a.u.

[0147] Thus, the advantages are a stronger MT-effect as compared to FIG. 6, the same MT-effect in a shorter duration as compared to FIGS. 7 and 8 and the same MT-effect with lower RF peak-amplitude as compared to FIG. 9, wherein no pre-saturation occurs compared to FIG. 10.

[0148] FIG. 16 shows a shifted dual-band Gaussian pulse similar to FIG. 15 with the difference that the pulse shift has been set to the minimum value for which the peak amplitude of the combined RF-pulse CP does not exceed the single-pulse maximum, e.g. by an algorithm as shown in FIG. 3. The duration is only 7 ms and the peak RF amplitude 10 a.u. with a pulse energy of 250 a.u. It has the same advantages as the combined RF-pulse CF of FIG. 15 over the prior art, but with a shorter duration.

[0149] Looking at the integer ratios of energy/duration in arbitrary units of the examples, wherein any ratio under 30 and any pre saturation is not desired (FIG. 10 omitted), one get:

TABLE-US-00007 Figure 6 7 8 9 15 16 energy E 125 250 250 250 250 250 Dur. T [ms] 5 10 10 5 7.5 7 E/T 25 25 25 50 34 36

[0150] Thus, compared to standard preparation schemes, the disclosed subject matter can generate similar MT-effects in a shorter duration while using moderate RF peak amplitudes.

[0151] FIG. 17 shows an optimized shifted dual-band Sinc RF-pulse with the two frequency offsets 380 Hz and 700 Hz and a dual-band rectangular frequency characteristic. In this example, the RF-pulse comprises two sub-pulses (initial RF-pulses P1, P2, see e.g. FIG. 2) with identical shapes (Sinc) and identical durations of 30 ms. The second sub-pulse starts 14 ms before the first sub-pulse ends. Due to their different frequency offsets, the overlap does not affect the frequency spectrum. With the selected pulse shift, the peak amplitude of the combined RF-pulse does not exceed the single-pulse maximum.

[0152] The envelope of the combined RF-pulse CP is:

[00005]$\begin{array}{cc}{B}_1\left(t\right)=A\left(\sin \left(\frac{t-t_0}{\sigma }\right)\frac{\sigma }{t-t_0}{e}^{{\mathrm{i\Delta \omega }}_{1}t}+A\sin \left(\frac{t-t_0-\Delta t}{\sigma }\right)\frac{\sigma }{t-t_0-\Delta t}{e}^{{\mathrm{i\Delta \omega }}_{2}t}\right).& \left(4\right)\end{array}$

[0153] The duration is only 46 ms and the peak RF amplitude 10 a.u. Thus, with the selected pulse shift, the peak amplitude of the combined RF-pulse does not exceed the single-pulse maximum.

[0154] Compared with the state of the art, the pulse duration is much lower than the 60 ms of the pulse shown in FIG. 11 while the peak RF amplitude is lower than of the pulse shown in FIG. 12 (20 a.u.).

[0155] Thus, the advantages are that the same saturation effect can be achieved in a shorter duration compared to FIG. 11 and the same saturation effect can be achieved with lower RF peak-power compared to FIG. 12.

[0156] FIG. 18 shows an RF-pulse with a dual-band non-Gaussian frequency characteristic. In this example, the RF-pulses comprises 6 sub-pulses (initial RF-pulses P1, P2, see e.g. FIG. 2) with identical Gaussian shapes and identical durations of 5 ms. The frequency offsets Δω/2π of the pulses are alternating between +1.2 kHz and −1.2 kHz for adjacent sub-pulses. Each succeeding sub-pulse starts 2000 μs before the preceding sub-pulse ends: this ensures that sub-pulses with identical frequency offset do not overlap, which preserves desirable frequency spectrum.

[0157] The duration is only 15 ms and the peak RF amplitude 10 a.u. with a pulse energy of 750 a.u. Thus, with the selected pulse shift, the peak amplitude of the combined RF-pulse does not exceed the single-pulse maximum.

[0158] Compared with the state of the art, the pulse energy corresponds with the pulse energies of FIGS. 13 and 14 with the advantage that the pulse duration is much lower than 30 ms. Thus, the advantages are that the same saturation effect can be achieved in a shorter duration compared to FIGS. 13 and 14.

[0159] Although the disclosed subject matter is in the form of preferred aspects and variations thereon, it will be understood that numerous additional modifications and variations could be made thereto without departing from the scope of the disclosed subject matter. For the sake of clarity, it is to be understood that the use of “a” or “an” throughout this application does not exclude a plurality, and “comprising” does not exclude other steps or elements. The mention of a “unit” or a “device” does not preclude the use of more than one unit or device.