Method for recording a magnetic resonance image data set, data carrier, computer-program product, and magnetic resonance system

Abstract

A method for recording a magnetic resonance image data set includes providing a magnetic resonance sequence. The magnetic resonance sequence includes at least one radio-frequency pulse and a slice-selection gradient pulse applied during or before the radio-frequency pulse, which is configured as non-constant. The method includes providing at least one correction term for compensating a magnetic field change of the slice-selection gradient pulse. The magnetic field change is ascertained via a transfer characteristic of the gradient system of the magnetic resonance system. The method also includes recording at least one magnetic resonance image data set with the magnetic resonance sequence using the correction term.

Claims

1. A method for recording a magnetic resonance image data set with a magnetic resonance system having a gradient system, the method comprising: providing a magnetic resonance sequence for performing a magnetic resonance scan, wherein the magnetic resonance sequence has at least one radio-frequency pulse and a slice-selection gradient pulse applied during or before the at least one radio-frequency pulse, wherein, at least for a predetermined time interval, the slice-selection gradient pulse is configured as non-constant during the application of the at least one radio-frequency pulse; providing at least one correction term for compensating a magnetic field change, wherein the magnetic field change is obtained as a change to an actual magnetic field compared to a setpoint magnetic field due to the slice-selection gradient pulse, wherein the magnetic field change is ascertained via a transfer characteristic of the gradient system of the magnetic resonance system, and the transfer characteristic represents the system characteristic of the gradient system in amplitude, phase, or amplitude and phase of different frequencies; recording at least one magnetic resonance image data set with the magnetic resonance sequence using the at least one correction term; and generating a corrected slice-selection gradient pulse using the at least one correction term.

2. In a non-transitory computer-readable storage medium that stores instructions executable by one or more processors to record a magnetic resonance image data set with a magnetic resonance system having a gradient system, the instructions comprising: providing a magnetic resonance sequence for performing a magnetic resonance scan, wherein the magnetic resonance sequence has at least one radio-frequency pulse and a slice-selection gradient pulse applied during or before the at least one radio-frequency pulse, wherein, at least for a predetermined time interval, the slice-selection gradient pulse is configured as non-constant during the application of the at least one radio-frequency pulse, wherein at least one gradient pulse of the magnetic resonance sequence leads to a magnetic field change to a basic magnetic field; providing at least one correction term for compensating a magnetic field change, wherein the magnetic field change is obtained as a change to an actual magnetic field compared to a setpoint magnetic field due to the slice-selection gradient pulse, wherein the magnetic field change is ascertained via a transfer characteristic of the gradient system of the magnetic resonance system, and the transfer characteristic represents the system characteristic of the gradient system in amplitude, phase, or amplitude and phase of different frequencies; recording at least one magnetic resonance image data set with the magnetic resonance sequence using the at least one correction term; and ascertaining a correction term for generating a compensation phase or a component of the compensation field, wherein the magnetic field change of the basic magnetic field is calculated as:
ΔB.sub.k,l=0(t)=.sup.1({Gnom,k(t)}.Math.GSTFk,l(f)}), and wherein l denotes the zero order of a transfer function of the gradient system of the magnetic resonance system, k denotes one of the gradient axes, is a Fourier transform, and G.sub.nom,k denotes the gradient pulse or pulses applied to the gradient axis.

3. A magnetic resonance system comprising: a gradient system; and a control facility configured to record a magnetic resonance image data set with a magnetic resonance system having a gradient system, the recordation of the magnetic resonance image data set comprising: provision of a magnetic resonance sequence for performing a magnetic resonance scan, wherein the magnetic resonance sequence has at least one radio-frequency pulse and a slice-selection gradient pulse applied during or before the at least one radio-frequency pulse, wherein, at least for a predetermined time interval, the slice-selection gradient pulse is configured as non-constant during the application of the at least one radio-frequency pulse; provision of at least one correction term for compensating a magnetic field change, wherein the magnetic field change is obtained as a change to an actual magnetic field compared to a setpoint magnetic field due to the slice-selection gradient pulse, wherein the magnetic field change is ascertained via a transfer characteristic of the gradient system of the magnetic resonance system, and the transfer characteristic represents the system characteristic of the gradient system in amplitude, phase, or amplitude and phase of different frequencies; recordation of at least one magnetic resonance image data set with the magnetic resonance sequence using the at least one correction term; and generation of a corrected slice-selection gradient pulse using the at least one correction term.

4. The method of claim 1, wherein the at least one correction term contains a current feed rating for gradient coils with which the corrected slice-selection gradient pulse is generated.

5. The method of claim 1, wherein the transfer characteristic has orders, and the magnetic field change is ascertained from at least one 1.sup.st order component.

6. The method of claim 1, wherein a transfer function is used as the transfer characteristic.

7. The method of claim 1, wherein the slice-selection gradient pulse is configured as at least temporarily non-constant during the application of the at least one radio-frequency pulse.

8. The method of claim 1, wherein the corrected slice-selection gradient pulse is generated via a pre-distortion filter for the gradient system of the magnetic resonance system.

9. The method of claim 1, wherein at least one gradient pulse of the magnetic resonance sequence leads to a magnetic field change to a basic magnetic field, and wherein the method further comprises ascertaining a correction term for generating a compensation phase or a component of the compensation field.

10. The method of claim 9, wherein the magnetic field change of the basic magnetic field is calculated as:
ΔB.sub.k,l=0(t)=.sup.1({Gnom,k(t)}.Math.GSTFk,l(f)}), and wherein l denotes the zero order of a transfer function of the gradient system of the magnetic resonance system, k denotes one of the gradient axes, is a Fourier transform, and G.sub.nom,k denotes the gradient pulse or pulses applied to the gradient axis.

11. The method of claim 9, wherein a phase deviation Df resulting from the magnetic field change is calculated as:
Df=∫g.Math.DB dt, wherein DB is the magnetic field change, g is the gyromagnetic ratio, and (∫ dt) is an integral over the time, and wherein a reference phase of the magnetic resonance system is adjusted by this value at predetermined points in time.

12. The method of claim 1, wherein the magnetic resonance system has a pre-distortion filter for the gradient system, and the pre-distortion filter is used to compensate deviations of all gradient pulses of the magnetic resonance sequence characterized by a Fourier transform.

13. The non-transitory computer-readable storage medium of claim 2, wherein the instructions further comprise generating a corrected slice-selection gradient pulse using the at least one correction term.

14. The non-transitory computer-readable storage medium of claim 13, wherein the at least one correction term contains a current feed rating for gradient coils with which the corrected slice-selection gradient pulse is generated.

15. The non-transitory computer-readable storage medium of claim 2, wherein the transfer characteristic has orders, and the magnetic field change is ascertained from at least one 1.sup.st order component.

16. The non-transitory computer-readable storage medium of claim 2, wherein a transfer function is used as the transfer characteristic.

17. The non-transitory computer-readable storage medium of claim 2, wherein the slice-selection gradient pulse is configured as at least temporarily non-constant during the application of the at least one radio-frequency pulse.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 shows one embodiment of a magnetic resonance system;

(2) FIG. 2 is a bSSFP sequence diagram with exemplary Cartesian k-space sampling;

(3) FIG. 3 shows one example of a two-dimensional radio-frequency pulse according to ee-classes.usc.edu/ee591/library/Pauly-RF.pdf;

(4) FIG. 4 shows exemplary k-space weighting and the spatial excitation profile of the two-dimensional pulse in FIG. 3;

(5) FIG. 5 shows an example of a spectral-spatial selective RF pulse, according to ee-classes.usc.edu/ee591/library/Pauly-RF.pdf;

(6) FIG. 6 shows the k-space weighting and the spin-echo profile of the pulse in FIG. 5;

(7) FIG. 7 shows an example of a transfer function for the zero order;

(8) FIG. 8 shows exemplary magnitude values of a transfer function for the 1st order (self-terms);

(9) FIG. 9 shows exemplary phase values of a transfer function for the 1st order (self-terms);

(10) FIG. 10 shows a first flow chart for recording a magnetic resonance data set; and

(11) FIG. 11 shows a second flow chart for recording a magnetic resonance data set.

DETAILED DESCRIPTION

(12) FIG. 1 shows one embodiment of a magnetic resonance system 1 with a scanner 2 and a control facility 3. The scanner 2 includes a gradient system 4 with three gradient coils 5, 6, and 7 for generating three gradient fields. The gradient fields that may be generated are aligned in the x, y, and z directions and are perpendicular to one another.

(13) Further, a transmitting coil arrangement 8 and a receiving coil arrangement 9 are arranged on the scanner 2. The transmitting coil arrangement 8 may be configured as a whole-body coil. The receiving coil arrangement 9 may be configured as a coil array. The receiving coil arrangement 9 is used as a local coil.

(14) In principle, however, the transmitting coil arrangement 8 may also be used to receive signals. However, the use of a receiving coil arrangement 9 not only increases the signal-noise ratio SNR per se, but the receiving coil arrangement 9 may also be used to perform parallel imaging. This may shorten the scan time.

(15) The control facility 3 of the magnetic resonance system 1 has a data carrier 10 on which a computer-program product 11 for performing the described method is saved. A magnetic resonance sequence 12 and different two-dimensional or spectral-spatial selective pulses 13, 14 may be stored on the data carrier 10. In addition, at least one pre-distortion filter p may be saved on the data carrier.

(16) Further, customary components of the magnetic resonance system 1, such as a patient couch, etc. are not shown for the sake of clarity.

(17) FIG. 2 shows a bSSFP sequence diagram 15 for the bSSFP scanning sequence 12 with a Cartesian sampling scheme. As usual, the imaging gradient axes are labeled G.sub.R for the read direction, G.sub.P for the phase-encoding direction, and G.sub.S for the slice selection direction. These may match the axes x, y, and z in FIG. 1, as described in the introduction but the imaging gradient axes do not have to.

(18) Even if the axes match, herein, fixed assignment is not mandatory; therefore, the axis of the read direction may lie in the direction of the x, y, or z axis. The axis of the phase-encoding direction and the axis of the slice selection direction are distributed between the remaining axes from the group of the x axis, y axis, and z axis.

(19) However, usually, the axes do not match. Then, the gradients in the directions G.sub.R, G.sub.P, and G.sub.S are formed by two or three of the gradient coils 5, 6, and 7 and thus of the gradients in the direction of the axes x, y, and z.

(20) HF denotes the axis for the radio-frequency pulses and acquisition windows.

(21) In order to excite only one slice with the radio-frequency pulses 16, a slice-selection gradient pulse 17 is applied simultaneously with the radio-frequency pulse 16 in the slice selection direction G.sub.S. The radio-frequency pulses 16 are configured as VERSE pulses. The profile of the slice-selection gradients 17 is configured accordingly. For purposes of comparison, a trapezoidal gradient pulse 18 is shown as a dotted line for a radio-frequency pulse with a sinc profile that selects the same slice. It can be seen that VERSE pulses require stronger but therefore shorter gradient pulses. The change in gradient strength is also very much greater when the gradient pulse 17 is switched than is the case with the gradient pulse 18. In order to compensate the dephasing effect of the slice-selection gradient pulse 17 on the magnetization in the transverse plane, the slice-selection gradient pulse 17 is followed directly by a slice-rephasing gradient pulse 19.

(22) The radio-frequency pulses of two successive sequence blocks may have alternating RF phases a and −a.

(23) A phase-encoding gradient pulse 20 is used in the phase-encoding direction G.sub.P. This is provided with varying strengths from sequence block to sequence block (e.g., in each of the n.sub.pe repetitions) in order to generate spatial encoding in the phase-encoding direction G.sub.P. This is indicated by the arrow 21.

(24) The read-dephasing gradient pulse 22 and the read-gradient pulse 23 are applied in the direction G.sub.R. Simultaneously with the read-gradient pulse 23, an acquisition window 24 is open in order to record the generated echo signal 25.

(25) In addition to these gradient pulses, which are present for imaging in all magnetic resonance sequences in which Cartesian k-space sampling is performed, a slice-dephasing gradient pulse 26 and a further slice-selection gradient pulse 17 is present in the slice selection direction G.sub.S. The slice-selection gradient pulse 17 is assigned to the next radio-frequency pulse 16 of the next sequence block. These provide that the bSSFP-scanning sequence 11 is “balanced” in the slice selection direction G.sub.S over a repetition time T.sub.R. This provides that the sums of the gradient moments in the slice selection direction G.sub.S are equal to zero over a repetition time T.sub.R.

(26) A phase-rewind gradient pulse 27 and a read-rewind gradient pulse 28 are also present for the same purpose. An arrow 29 with the reverse direction to that of the arrow 21 is arranged in the phase-rewind gradient pulse 27. This shows the opposite polarity of the phase-encoding gradient pulse 20 compared to the phase-rewind gradient pulse 27 with the same strength in each case.

(27) Thus, the gradient pulses in the read direction G.sub.R and slice selection direction G.sub.S are arranged symmetrically with respect to the point in time 30; the phase-encoding gradient pulse 20 and the phase-rewind gradient pulse 27 are arranged antisymmetrically. The gradient pulses are thus arranged symmetrically in terms of time with respect to the point in time 30.

(28) A sequence block extends from one radio-frequency pulse to the next and has the length TR. The excitation section includes the excitation pulse 16 and the slice-selection gradient pulse 17. The detection section includes the read-gradient pulse 23. While this is applied, an acquisition window is also open.

(29) The bSSFP scanning sequence 12 is only shown as an example of a scan sequence in which a VERSE pulse is used.

(30) FIG. 3 shows a multi-dimensional radio-frequency pulse in the form of a two-dimensional radio-frequency pulse 31 (e.g., 2D spatial pulse) together with an associated slice-selection gradient pulse 32. The slice-selection gradient pulse 32 uses gradients in two directions, Gx and Gy, where Gy is 90° out of phase with Gx in order to obtain a spiral k-space trajectory. This is depicted at the top of FIG. 4. Hence, an approximately cuboidal volume may be selected with the 2D spatial pulse 31, where the excitation profile of the pulse 31, 32 in the x and y directions is depicted at the bottom of FIG. 4. Therefore, the selection takes place in two spatial dimensions. Herein, the slice-selection gradient pulse 32 is configured as oscillating. With this shape of a slice-selection gradient pulse, k-space may be traversed during the application of the radio-frequency pulse 31, and thus, the two-dimensional selection may be achieved.

(31) FIG. 5 shows a multi-dimensional radio-frequency pulse in the form of a spectral-spatial-selective radio-frequency pulse 33 together with an associated slice-selection gradient pulse 34. The RF pulse 33 consists of many contiguous sub-pulses, while the slice-selection gradient pulse 34 is configured as oscillating in a sawtooth-like manner (e.g., switches rapidly back and forth from positive to negative gradients). Herein, a short plateau is maintained at the positive and negative maximum in each case during which an RF sub-pulse is played-in. The spatial profile of the pulse is defined by the shape of the single sub-pulses, and the spectral profile is defined by the envelope curve of the sub-pulse amplitudes. In this way, a spectral component and a slice are selected simultaneously.

(32) The setpoint gradient field 34 is depicted as a solid line in FIG. 5. However, without the correction according to the present embodiments, the gradient field 34′ shown as dotted lines would be played out; this has less precisely defined corners and also lags slightly behind the setpoint gradient field 34. Therefore, without the correction terms according to the present embodiments, the gradient pulse 34′ would not be optimally matched to the RF pulse 33, and artifacts would occur. The at least one correction term determined by the gradient characteristic enables the difference between the setpoint gradient field 34 and the actual gradient field 34′ to be compensated by a compensation field ΔG or the corrected gradient pulse 34 to be output immediately instead of the gradient pulse 34′.

(33) FIG. 6 further shows the k-space weighting and the spin-echo profile of the spectral-spatial pulse in FIG. 5 for purposes of illustration.

(34) Herein, the slice-selection gradient pulses 17, 32, and 34 are in each case corrected according to the present embodiments.

(35) The radio-frequency pulses 31 or 33 may, in each case with slice-selection gradients 32 or 34, respectively, be used in any scanning sequence in order to enable a different selection instead of a slice.

(36) It is common to all slice-selection gradients 17, 32, and 34 that the slice-selection gradients 17, 32, and 34 are non-constant.

(37) The transfer function GSTF used to ascertain the magnetic field change to be corrected with respective components GSTF.sub.x, GSTF.sub.y and GSTF.sub.z (e.g., the cross-terms for the zero order) is depicted with a corresponding magnitude component in FIG. 7. The frequency in kHz is plotted on the axis 42, and a standardized magnitude value is plotted on the axis 43. Line 44 shows the component in the x direction; line 45 shows the component in the y direction, and line 46 shows the component in the z direction. The components may be measured, for example, using the field camera method.

(38) The formula presented above
ΔB.sub.k,l=0(t)=.sup.1({Gnom,k(t)}.Math.GSTFk,l(f)}).
may be used to ascertain a change to the magnetic field DB.sub.0,x, DB.sub.0,y, and DB.sub.0,z from the transfer function GSTF or corresponding components GSTF.sub.x, GSTF.sub.y, and GSTF.sub.z in each case. This results in a phase change Df.

(39) FIG. 8 shows the magnitude component of the components GSTF.sub.xx, GSTF.sub.yy, and GSTF.sub.zz of the transfer function GSTF, and thus the 1st order self-terms. Once again, the frequency in kHz is plotted on the axis 42, and a standardized magnitude value is plotted on the axis 43. Herein, it may be seen that the gradient system is configured as a low-pass filter: while low frequencies are transmitted almost in a ratio of 1:1, the proportion decreases sharply toward higher frequencies.

(40) Line 47 shows the component GSTF.sub.xx, line 48 shows the component GSTF.sub.yy, and line 49 shows the component GSTF.sub.zz. These components may again be measured using the field camera method.

(41) FIG. 9 shows the phase component of the components GSTF.sub.xx, GSTF.sub.yy, and GSTF.sub.zz of the transfer function GSTF and thus of the 1st order. Once again, the frequency in kHz is plotted on the axis 42, and a standardized phase value is plotted on the axis 50.

(42) Line 51 shows the phase values of the component GSTF.sub.xx, line 52 shows the phase values of the component GSTF.sub.yy, and line 53 shows the phase values of the component GSTF.sub.zz. These components may again be measured using the field camera method.

(43) The components shown in FIGS. 8 and 9 may be used to correct the GIRF-based artifacts of the gradient pulses or gradient axes with respect to one another. The zero order according to FIG. 7 may be used to ascertain the phase deviation.

(44) FIG. 10 shows a first flow chart for recording a magnetic resonance image data set.

(45) In act S1, the pre-distortion filter p is provided. The pre-distortion filter was ascertained using a transfer function GSTF or, to be more precise, at least the 1st order of the components GSTF.sub.xx, GSTF.sub.yy, and GSTF.sub.zz.

(46) In act S2, a magnetic resonance sequence (e.g., a magnetic resonance sequence including a two-dimensional slice-selection pulse 12 or a spectral-spatial selective pulse 13) is provided.

(47) Usually, after a magnetic resonance sequence has been loaded, parameters such as the position of the slice or slices, the resolution, the number of image elements, etc. are set in act S3. Only then is the exact configuration of the gradients fixed.

(48) The sequence of gradients or gradient pulses resulting from this magnetic resonance sequence 12 may either be used as a whole in order to calculate pre-distorted gradients G.sub.pre taking into account all effects using the pre-distortion filter p in act S4. Herein, pre-distorted gradient strengths that generate additional compensation fields when played out are ascertained. The effects taken into account are at least the self-terms and, for example, the 1st order cross-terms. Then, in act S5, a magnetic resonance image data set is recorded with the magnetic resonance sequence. This may be a 2D image data set with one or more slices, a 3D image data set, or even a 4D image data set.

(49) FIG. 11 shows a second embodiment of a flow chart for recording a magnetic resonance data set. After acts S1 to S3, the provision of the transfer function and the magnetic resonance sequence and the setting of the parameters of the magnetic resonance sequence, this has the act S6. In this act, the gradient pulses are collectively sent in sections (e.g., for a repetition time T.sub.R) as a digital gradient sequence signal to a control facility, a pre-distortion filter p is applied thereto, and a resultant signal is transmitted as a pre-distorted gradient signal G.sub.pre1 to the magnetic resonance system.

(50) Herein, the pre-distortion filter p is simple to implement since, as described above, nominal gradient pulses G.sub.nom are converted into pre-distorted gradient pulses G.sub.pre from short sections.

(51) Repeated calculation is necessary since at least the phase-encoding gradient pulse 20 and then also the phase-rewind gradient pulse 27 have different values in different sequence blocks. This embodiment may, for example, be used in the case of sampling with continuous sequence blocks.

(52) The elements and features recited in the appended claims may be combined in different ways to produce new claims that likewise fall within the scope of the present invention. Thus, whereas the dependent claims appended below depend from only a single independent or dependent claim, it is to be understood that these dependent claims may, alternatively, be made to depend in the alternative from any preceding or following claim, whether independent or dependent. Such new combinations are to be understood as forming a part of the present specification.

(53) While the present invention has been described above by reference to various embodiments, it should be understood that many changes and modifications can be made to the described embodiments. It is therefore intended that the foregoing description be regarded as illustrative rather than limiting, and that it be understood that all equivalents and/or combinations of embodiments are intended to be included in this description.