Method for reconstructing a single slice image data set from a k-space measurement data set

Abstract

According to the present invention, a method for reconstructing a single slice image data set from k-space measurement data sets recorded simultaneously from at least two slices of an object under examination comprises iteratively modifying and thereby optimizing a k-space trajectory based on which a single image data set is reconstructed from measurement data sets recorded overlappingly from a plurality of slices, thereby retrospectively reducing interference in the obtained single image data set.

Description

Technical Field

[0001] The invention relates to a method for reducing artifacts in an image data set recorded by means of magnetic resonance, in particular to a method for improved reconstruction of a single-slice image data set from a k-space measurement data set, which was recorded simultaneously from at least two slices of an examination object. Background Art

[0002] Magnetic resonance technology (hereinafter the abbreviation MR stands for magnetic resonance) is a known technology, with the help of which images of the interior of the object under examination can be generated. In short, for this purpose, the object under examination is positioned in a relatively strong static, uniform basic magnetic field, also called B0 field, with a field strength of 0.2 Tesla to 7 Tesla and more in a magnetic resonance device, so that the nuclear spins of the object under examination are oriented along the basic magnetic field. In order to trigger the nuclear spin resonance that can be measured as a signal, a radio frequency excitation pulse (RF pulse) is irradiated into the object under examination, the triggered nuclear spin resonance is measured as so-called k-space data and MR images are reconstructed or spectroscopic data are determined based on the k-space data. In order to positionally encode the measurement data, a rapidly switched magnetic gradient field, referred to as gradient, is superimposed on the basic magnetic field. The used pattern (schema) that describes the time sequence of the RF pulses to be irradiated and the gradients to be switched is called a pulse sequence (schema), or also referred to as a sequence. The recorded measurement data are digitized and stored as complex values ​​in a k-space matrix. From the k-space matrix occupied by the values, the associated MR image can be reconstructed, for example, by means of a multidimensional Fourier transform.

[0003] Probably the most commonly used method for generating echo signals after excitation of nuclear spins is the so-called spin echo method. In the simplest case, it can be said that the transverse magnetization is "deflected" by irradiating at least one RF refocusing pulse after the RF excitation pulse, whereby the dephased magnetization is phased again and thus a so-called spin echo SE is generated after a time TE called the echo time after the RF excitation pulse.

[0004] The excitation and measurement of the generated echo signals are repeated after a repetition time TR (eg, with switching of different gradients for spatial encoding) until the required number of echo signals has been measured and stored in k-space in order to be able to image the examination object.

[0005] Under the SE sequence, in particular the TSE sequence (TSE: “Turbo Spin Echo”), which is also known under the name FSE (“Fast Spin Echo”) or RARE (“Rapid Acquisition with Refocused Echoes”) sequence, is widely popular in clinical applications. Compared with a “simple” SE sequence, the advantage of the TSE sequence is that a plurality of refocusing pulses are switched after the RF excitation pulse, and thus a plurality of spin echo signals SE are also generated after the excitation. This speeds up the data acquisition, since fewer repetitions of the sequence with different spatial encodings are required in order to measure all the desired data. In the case of a TSE sequence, compared with conventional SE methods, the measurement time for the entire k-space is reduced by a so-called “fast factor”, which corresponds to the number of echo signals that are refocused and recorded after the excitation.

[0006] The desire for faster and faster MR recordings in the clinical environment is currently leading to a revival of methods for simultaneously recording multiple images. In general, these methods can be characterized in that, at least during a portion of the measurement, the transverse magnetization of at least two slices is used simultaneously in a targeted manner for the imaging process ("multi-slice imaging", "slice multiplexing"). In contrast, in the established "multi-slice imaging", the signals of at least two slices are recorded alternately, i.e. completely independently of one another, with correspondingly longer measurement times.

[0007] Known methods for this purpose are, for example, so-called Hadamard encoding, methods with simultaneous echo refocusing, methods with broadband data recording or also methods using parallel imaging in the slice direction. The last-mentioned methods also include, for example, the CAIPIRINHA technique, as described by Breuer et al. in “Controlled Aliasing in Parallel Imaging Results in Hiher Acceleration (CAIPIRINHA) for Multi-Slice Imaging”, Magnetic Resonance in Medicine 53, 2005, pp. 684-691, and the blipped CAIPIRINHA technique, as described by Setsompop et al. in “Blipped-Controlled Aliasing in Parallel Imaging for Simultaneous Multislice Echo Planar Imaging With Reduced g-Factor Penalty”, Magnetic Resonance in Medicine 67, 2012, pp. 1210-1224.

[0008] In such slice multiplexing methods, so-called multi-band RF pulses are used in order to simultaneously excite or manipulate, for example refocus or saturate, two or more slices. For example, such a multi-band RF pulse is a multiplexing of individual RF pulses which can be used to manipulate the individual slices to be manipulated simultaneously. In order to be able to separate the generated signals of different slices, for example, before the multiplexing, a different phase is applied to the individual RF pulses, for example by adding a linear phase increase, so that the slices are moved relative to each other in the position space. For example, by the multiplexing, a baseband-modulated multi-band RF pulse is obtained from the addition of the pulse shapes of the individual RF pulses.

[0009] As described, for example, in the article by Setsompop et al. mentioned above, the g-factor loss caused by the offset between slices can be reduced, for example, by using gradient blips or by correspondingly modulating the phase of the individualized RF pulses. As also described in the article by Setsompop et al. mentioned above, but also in the article by Breuer et al. mentioned above, the signals of slices excited or manipulated in another way can first be combined like the signal of only one slice in order to be separated later in post-processing by a parallel reconstruction method, for example the (slice) GRAPPA method (GRAPPA: “Generated Autocalibrating Partial Parallel Acquisition”), or another parallel imaging method (PPA), for example the SENSE method (SENSE: Sensitivity encoding).

[0010] For such a separation of overlapping recorded slices, individually recorded reference data are usually used for each of the slices, which reference data have been measured, for example, in a pre-scan.

[0011] If tomomultiplexing methods with the aforementioned gradient spikes and, for example, TSE sequences are used, in recordings in which the spin signal in at least one tissue type is not suppressed, for example in non-fat-saturated recordings, ghosting artifacts may be observed in the generated images. Summary of the invention

[0012] The invention is based on the object of reducing artifacts in tomosynthesis methods, in particular without fat saturation.

[0013] The invention is based on the recognition that ghosting artifacts may arise due to different effects of the gradient peaks used on spins in different tissue types. Due to differences in the resonance frequencies of the spins present in different tissues, for example in water and fat tissue, which are called chemical shifts, the spins in the different tissue types are not excited in the same slice, but in slices that are offset relative to each other by the slice distance Δz. As a result, the gradient peaks used also induce different phase shifts in the slice direction for the spins bound in the different tissue types, also called field of view shifts (English “FOV-shifts”, FOV: “field of view”). The gradient peaks are usually selected so that they cause the desired phase shift for the spins bound in water, but they thus cause a different phase shift for the spins bound in fat, for example. However, there is also a dependence of the phase shifts of the spins in all tissue types on the gradient moments of the applied gradient peaks.

[0014] Furthermore, other undesirable effects that impair the magnetization of the spins can influence the signal intensity present during the switched gradient spikes, in particular since gradient spikes with different gradient moments can be used during the recording. Such effects can include, for example, eddy currents, (already slight) violations of the CPMG conditions (CPMG: Carr-Purcell-Meiboom-Gill) or different signal intensities of the resulting FID signals (FID: free induction decay), which are caused, for example, by different signal crushing effects (English: "signal-crushing effects") that can also cause gradient spikes.

[0015] Furthermore, ghosting artifacts may be intensified by differences in the recording parameters used during imaging on the one hand and during recording of reference data for tomosynthesis on the other hand, such as slice profiles, bandwidth of the RF pulses, flip angles, relaxation times.

[0016] The object is achieved by a method according to an embodiment for reconstructing a single slice image data set from k-space measurement data sets acquired simultaneously from at least two slices of an examination object, a magnetic resonance apparatus according to an embodiment, a computer program according to an embodiment and an electronically readable data carrier according to an embodiment.

[0017] The method according to the invention for reconstructing a single slice image data set from k-space measurement data sets acquired simultaneously from at least two slices of an examination object comprises the following steps:

[0018] a) loading a k-space measurement data set (MD) comprising measurement data acquired simultaneously for at least two slices of the examination object along a defined k-space trajectory,

[0019] b) reconstructing a reference slice image data set for each of the at least two slices from the k-space measurement data set,

[0020] c) determining a test region for the reference tomographic image data set, in which no artifacts are to be expected,

[0021] d) determining a reference value for each of the at least two slices by analyzing a test region in a corresponding reference slice image data set,

[0022] e) modifying at least one of the parameters characterizing the prescribed k-space trajectory,

[0023] f) reconstructing a test slice image data set for each of the at least two slices from the k-space measurement data set using the modified parameters,

[0024] g) determining a test reference value corresponding to the reference value for each of the at least two slices by analyzing a test region in the corresponding test slice image data set,

[0025] h) repeating steps e) to g) with a modification different from the one already tested until the abort criterion is reached,

[0026] i) compare the reference value with the test reference value according to the quality standard,

[0027] j) storing and comparing the reference value (R1, ..., Rn) and the test reference value (R1 i *, …, Rn i *) in the optimal value of the associated parameter (P i ).

[0028] The stored parameters can now be used to reconstruct slice image data sets of individual slices from the k-space measurement data set, in which slice image data sets artifacts are reduced compared to the reference slice image data set.

[0029] The method allows retrospective reduction of artifacts, in particular slice cross-talk artifacts, and thus improves the image quality in image data sets of individual slices acquired using slice multiplexing, without having to extend the measurement time required for acquiring the measurement data.

[0030] The magnetic resonance system according to the present invention comprises a magnet unit, a gradient unit, a radio frequency unit and a control device which is designed to carry out the method according to the present invention and has a parameter modification unit.

[0031] When the computer program according to the invention is executed on a control device, it implements the method according to the invention on the control device.

[0032] The computer program may also be in the form of a computer program product which can be directly loaded into a memory of a control device and has program code means for carrying out the method according to the invention when the computer program product is executed in a computing unit of a computing system.

[0033] The electronically readable data carrier according to the invention comprises electronically readable control information stored thereon, which comprises at least one computer program according to the invention and is designed to execute the method according to the invention when the data carrier is used in a control device of a magnetic resonance system.

[0034] The advantages and embodiments described with reference to the method also apply analogously to the magnetic resonance system, the computer program product and the electronically readable data carrier. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Further advantages and details of the present invention are obtained from the embodiments described below and from the accompanying drawings. The examples listed are not intended to limit the present invention. The accompanying drawings show:

[0036] Figure 1 A schematic flow chart of the method according to the present invention is shown,

[0037] Figure 2 A schematically illustrated magnetic resonance system according to the invention is shown. DETAILED DESCRIPTION

[0038] Figure 1 is a schematic flow chart of a method according to the invention for reconstructing a single slice image data set from a k-space measurement data set which was acquired simultaneously from at least two slices of an examination object.

[0039] Here, a k-space measurement data set MD is loaded, which includes measurement data that were recorded simultaneously for at least two slices S1, S2, ..., Sn of the examination object along a defined k-space trajectory, for example along a k-space line. Here, the at least two slices S1, S2, ..., Sn can be n slices out of a total of N (N≥n) slices to be examined of the examination object. The simultaneous recording of the measurement data can be achieved by using slice multiplexing techniques, in particular by using gradient spikes for applying the desired phase shift.

[0040] A reference slice image dataset B1, ..., Bn is reconstructed for each of at least two slices S1, S2, ..., Sn from the k-space measurement dataset MD, the reference slice image dataset B1, ..., Bn comprising the image data of the corresponding slice (block 101). To this end, the measurement data recorded overlappingly along the k-space trajectory for at least two slices S1, S2, ..., Sn are separated into the data of the individual slices, for example by means of a slice GRAPPA method or optionally by means of a Fourier transform. The reconstruction can usually be carried out in accordance with the slice multiplexing technique used when recording the measurement data. The k-space trajectory used reflects the desired spatial encoding of the recorded measurement data.

[0041] A test region Rt is determined for the reference tomographic image data sets B1, ..., Bn in which no artifacts are expected (block 103).

[0042] The test regions Rt of different reference slice image datasets B1, ..., Bn can be different. Thus, an individually adapted test region Rt can be selected for each reference slice image dataset B1, ..., Bn. Alternatively, a test region Rt can also be sought for which no artifacts are expected in all reference slice image datasets B1, ..., Bn. It is conceivable that a correspondingly trained neural network, which has optionally already been used for quality assurance, determines the test region in that the neural network evaluates the anatomical region, for example, with respect to its artifact intensity.

[0043] The test region Rt can be selected as a region of the corresponding reference slice image data set B1, ..., Bn in which the examination object is not to be imaged. In this way, the test region Rt can be easily determined, for example by means of a segmentation method which segments the examination object.

[0044] If, in addition to the k-space measurement data MD, reference data RD are also recorded for each of the at least two slices S1, S2, ..., Sn, for example for calibration or for determining coil sensitivities for reconstructing the image data set, the determination of the test region Rt can be carried out in the image data of the slices S1, S2, ..., Sn reconstructed from the reference data RD. In particular, when the reference data RD are recorded with a lower resolution, the determination of the test region can be accelerated by a smaller amount of data. Furthermore, the reference data RD can also be obtained from a subset of the measurement data MD recorded for imaging. For example, the central k-space line can be used in order to obtain a low-resolution data set.

[0045] The determination of the test region Rt can take into account the field of view offset (FOV offset) used when recording the k-space measurement data set MD. In particular, the association of the test region with one of the slices can be adjusted corresponding to the FOV offset. This is particularly advantageous in the following case: a FOV offset>1 is used and the test region is determined in image data reconstructed from reference data. If the measurement data recorded overlappingly along the k-space trajectory for at least two slices S1, S2, ..., Sn are separated into data of each slice by means of Fourier transformation, then the test region should be a non-overlapping region of at least two simultaneously recorded slices S1, S2, ..., Sn.

[0046] The test region Rt may be stored, for example, in the form of a mask which, when applied to the image data set, only still allows viewing of the test region in the image data set.

[0047] For each of the at least two slices S1, S2, ..., Sn, a reference value R1, ..., Rn is determined in each reference slice image data set B1, ..., Bn by analyzing the test region Rt, in particular analyzing the signal strength present in the test region Rt (block 105). In this case, the reference value is selected in particular such that it correlates with the presence of disturbances, such as artifacts, in the image.

[0048] For example, the reference values ​​R1, ..., Rn of the test region Rt can reflect the signal strength present in the test region, in particular the average signal strength in the test region Rt. The signal strength can be easily determined, for example also via an application using a correspondingly trained neural network, and compared with an expected value. In particular, if the test region Rt is selected such that no examination object is to be imaged in the test region Rt and therefore no signal strength is expected, the signal strength still present in the test region indicates the presence of interference.

[0049] In a further step, the parameter P characterizing the used k-space trajectory is modified. iAt least one parameter in (block 107). The spatial encoding assumed for the recorded measurement data is thereby changed.

[0050] The parameter P that characterizes the specified k-space trajectory i For example, these may be gradients applied during the acquisition of the measurement data of the k-space measurement data set MD, in particular the amplitude of the readout gradient, and / or phases applied during the acquisition of the measurement data of the k-space measurement data set MD.

[0051] Parameter P i The modification of may be a change in a constant value and / or a change according to a desired function, for example a linear slope.

[0052] It is conceivable here that when modifying the parameters of the k-space trajectory used, different parameters P i For example, in the modification of a k-space line used as a k-space trajectory, the amplitude of the readout gradient can be changed at a constant value, and the phase applied during recording can be changed by adding a linear slope.

[0053] Using the modified parameter P i , reconstructing a test slice image dataset B1 for each of at least two slices S1, S2, ..., Sn from the k-space measurement dataset MD i *, ..., Bn i *(Block 101'). Here, in addition to using the modified parameter P i And thus assuming that, except for the changed spatial encoding, a similar procedure is performed as in block 101.

[0054] For each of the at least two slices S1, S2, ..., Sn, by analyzing the corresponding test slice image data set B1 i *, ..., Bn i * to determine the test reference value R1 corresponding to the reference value R1, ..., Rn i *, …, Rn i *(Block 105'). The test reference value R1 corresponding to the reference values ​​R1, ..., Rn can be easily obtained. i *, …, Rn i *, in that the test slice image data set B1 in block 105 ′ is analyzed similarly to the reference image data sets B1 , . . . , B2 in block 105 i *, ..., Bn i *.

[0055] According to the quality standard, the reference values ​​R1, ..., Rn are compared with the test reference value R1 i *, …, Rn i* is compared (block 109). In particular, for at least one of the at least two faults S1, S2, ..., Sn, the associated reference value can be compared with at least one test reference value corresponding to the same fault. Advantageously, the comparison is performed for all faults of the at least two faults S1, S2, ..., Sn. It is conceivable that the value (R1, ..., Rn or R1i*, ..., Rni*) that is better according to the comparison is retained as the applicable new reference value R1, ..., Rn so that the next test reference value is compared with the applicable new reference value. At least the value corresponding to the comparison between the reference value (R1, ..., Rn) and the test reference value (R1) is stored. i *, …, Rn i *) is the parameter P associated with the best value i .

[0056] The quality criteria describe the degree of freedom from artifacts and can achieve reference values ​​R1, ..., Rn and test reference values ​​R1 i *, …, Rn i If the reference value is associated with a disturbance in the image, the quality criterion may be selected such that it allows an assessment whether the reference value or a test reference value compared to the reference value is associated with a stronger disturbance and returns a value associated with a smaller disturbance as a better value.

[0057] For example, if the reference values ​​R1, ..., Rn of the test region Rt reflect the signal strength, the quality criterion may require the smallest possible signal strength in the test region. A comparison of the reference value and the associated test reference value may identify the value with the smaller signal strength as the better value.

[0058] With the aid of at least one parameter P i Blocks 107, 101' and 105' are repeated for each modification that is different from the modification that has already been tested until the termination criterion is reached (query 111). The comparison can also be repeated for the applicable reference value and the test reference value (block 109) in this case. However, it is also conceivable in a simple embodiment to perform a comparison between the existing reference data and the test reference data only after all desired modifications have been performed.

[0059] The modifications performed in the different processes of block 107 can, for example, change at least one parameter P by a preset value in steps of increasing or decreasing or according to an optimization algorithm, such as a greedy algorithm, in particular a golden section search, or a binary search algorithm. iThe implementation of the preset values ​​already achieves good results and defines the time and computing effort required for the iterations. Using an optimization algorithm can lead to better results, but also increases the time and computing effort or at least makes it less predictable.

[0060] The termination criterion can take into account a preset maximum number of executions of block 107 (counter i) and / or a minimum threshold value for the result of the comparison performed of the reference values ​​R1, ..., Rn with the test reference values ​​R1i*, ..., Rni*. For example, the termination criterion can be met after different modifications performed of the desired number Z. Additionally or alternatively, the termination criterion can be met if the comparison results in that all test reference values ​​show a negligible improvement in any case compared to the applicable reference value. What is considered negligible can be selected depending on the application. The larger the value that is considered negligible, the faster the termination criterion is reached, and the smaller the value that is considered negligible, the higher the quality improvement that can be achieved with the aid of the method.

[0061] If the termination criterion is met (query 111, rightward arrow), the parameter P can be used to i The slice image data sets B1', ..., Bn' are reconstructed from the k-space measurement data set MD. The parameters P i According to the comparison between the reference values ​​R1, ..., Rn and the test reference value R1 i *, …, Rn i * is associated with a reduction in disturbances, such as ghosting artifacts, in the tomographic image data set.

[0062] The method described in this article allows the k-space trajectory based on which the individual image data sets are reconstructed to be optimized via iterative modification and thus to retrospectively reduce interference in the acquired individual image data sets.

[0063] If the method is executed for a tuple consisting of at least two slices S1, S2, . . . , Sn to be recorded simultaneously out of a total of N (N>n) slices to be recorded, the parameter P obtained in one execution is i It can also be used when executing another tuple in order to achieve a reduction in interference. This can save computing effort and computing time. For example, if two slices are recorded simultaneously for each tuple and a total of eight slices are to be measured, the parameters determined according to the invention for one of the 2-tuples can also be used for the other three 2-tuples. Alternatively, a separate parameter P can also be determined for each tuple. i , whereby the reduction of artifacts can be increased.

[0064] Figure 2The magnetic resonance apparatus 1 according to the present invention is schematically shown and comprises a magnet unit 3 for generating a basic magnetic field, a gradient unit 5 for generating gradient fields, a radio frequency unit 7 for radiating and receiving radio frequency signals and a control device 9 configured to execute the method according to the present invention.

[0065] exist Figure 2 1 and 2. These subunits of the magnetic resonance device 1 are only roughly schematically shown in FIG. In particular, the radio frequency unit 7 can be composed of a plurality of subunits, for example a plurality of coils, such as the schematically shown coils 7.1 and 7.2 or more coils, which can be designed either only for transmitting radio frequency signals or only for receiving triggered radio frequency signals or for both.

[0066] To examine an examination subject U, for example a patient or also a phantom, the examination subject can be introduced into the magnetic resonance system 1 on an examination couch L in a measurement volume of the magnetic resonance system 1. Slices Sa and Sb represent by way of example slices of the examination subject to be recorded simultaneously, from which echo signals are to be recorded and detected as measurement data.

[0067] The control device 9 serves to control the magnetic resonance system 1 and can in particular control the gradient unit 5 by means of a gradient control 5 ′ and the radio frequency unit 7 by means of a radio frequency transmission / reception control 7 ′. The radio frequency unit 7 can include a plurality of channels on which signals can be transmitted or received.

[0068] The radio frequency unit 7 together with its radio frequency transmission / reception control device 7' is responsible for generating and irradiating (transmitting) a radio frequency alternating field, which is used to manipulate the spins in the area to be manipulated (for example, in the slice S to be measured) of the examination object U. Here, the intermediate frequency of the radio frequency alternating field, which is also called the B1 field, is usually set as close as possible to the resonance frequency of the spins to be manipulated. The deviation of the intermediate frequency from the resonance frequency is called off-resonance. In order to generate the B1 field, a controlled current is applied to the HF coil in the radio frequency unit 7 with the help of the radio frequency transmission / reception control device 7'.

[0069] Furthermore, the control device 9 comprises a parameter modification unit 15, by means of which the parameters according to the invention can be modified. The control device 9 is designed overall for carrying out the method according to the invention.

[0070] The calculation unit 13 included by the control device 9 is designed to perform all the calculation operations required for the necessary measurements and determinations, in particular also for the comparison and determination according to the invention of the reference values ​​to be retained and the associated parameters. Intermediate results and results required for this or determined in this way can be stored in a memory unit S of the control device 9. The units shown here do not necessarily mean physically separated units, but only represent a division into meaningful units, which can, for example, also be realized with fewer physical units or also with only a single physical unit.

[0071] Via the input / output device E / A of the magnetic resonance system 1 , control commands can be supplied to the magnetic resonance system, for example by a user, and / or results of the control device 9 , for example image data, can be displayed.

[0072] The method described herein may also be present in the form of a computer program product, which includes a program and, when the program is executed on the control device 9, implements the described method on the control device 9. There may also be an electronically readable data carrier 26 with electronically readable control information stored thereon, which includes at least one computer program product of the type just described and is designed such that, when the data carrier 26 is used in the control device 9 of the magnetic resonance system 1, the electronically readable control information executes the described method.

Claims

1. A method for reconstructing a single slice image data set from a k-space measurement data set (MD), the k-space measurement data set (MD) being recorded simultaneously from at least two slices (S1, S2, ..., Sn) of an examination object, the method comprising the following steps: a) loading a k-space measurement data set (MD) comprising measurement data recorded simultaneously for at least two slices (S1, S2, . . . , Sn) of the examination object along a defined k-space trajectory, b) reconstructing a reference slice image data set (B1, . . . , Bn) for each of the at least two slices (S1, S2, . . . , Sn) from the k-space measurement data set (MD), c) determining a test region (Rt) for the reference tomographic image data set (B1, . . . , Bn) in which no artifacts are to be expected, d) determining a reference value (R1, ..., Rn) for each of the at least two slices (S1, S2, ..., Sn) by analyzing the test region in the corresponding reference slice image data set (B1, ..., Bn), e) Modify the parameters (P) that characterize the specified k-space trajectory i ), f) Using the modified parameters (P i ) reconstructs a test slice image data set (B1) for each of the at least two slices (S1, S2, . . . , Sn) from the k-space measurement data set (MD); i *,…,Bn i *), g) by analyzing the corresponding test tomographic image data set (B1 i *,…,Bn i *), determining a test reference value (R1) corresponding to the reference value (R1, ..., Rn) for each of the at least two faults (S1, S2, ..., Sn) i *,…,Rn i *), h) repeating steps e) to g) with a modification different from the one already tested until the abort criterion is reached, i) comparing the reference value (R1, ..., Rn) and the test reference value (R1) according to the quality standard i *,…,Rn i *) for comparison, j) storing and comparing the reference value (R1, . . . , Rn) with the test reference value (R1) i *,…,Rn i *) in the optimal value of the associated parameter (P i ).

2. The method according to claim 1, wherein the test regions (Rt) of different reference tomographic image data sets (B1, . . . , Bn) can be different.

3. The method according to claim 1 or 2, wherein a test region (Rt) is a region in the reference tomographic image data set (B1, . . . , Bn) in which the examination object is not to be imaged.

4. A method according to claim 1 or 2, wherein in addition to the k-space measurement data set (MD), reference data (RD) are also recorded for each of the at least two slices, and the determination of the test area (Rt) is carried out in the image data reconstructed from the reference data.

5. The method according to claim 1 or 2, wherein determining the test region (Rt) takes into account a field of view offset used when recording the k-space measurement data set (MD).

6. The method according to claim 1 or 2, wherein the parameter (P i ) is the amplitude of the gradient applied during the recording of the measurement data of the k-space measurement data set (MD) and / or the phase applied during the recording. The method of claim 6 , wherein the k-space trajectory is a k-space line. The method of claim 6 , wherein the gradient is a readout gradient.

9. The method according to claim 1 or 2, wherein the parameter (P i ) is modified by a change in a constant value.

10. The method according to claim 1 or 2, wherein the parameter (P i ) is modified according to the change of the desired function. The method of claim 10 , wherein the desired function is a linear slope.

12. The method according to claim 1 or 2, wherein the modification performed during the different steps of step e) gradually changes at least one parameter (P i ).

13. The method according to claim 1 or 2, wherein the modification performed in the different processes of step e) changes at least one parameter (P i ).

14. The method according to claim 1 or 2, wherein the modification performed in the different processes of step e) changes at least one parameter (P i ).

15. The method according to claim 1 or 2, wherein the modification performed in the different processes of step e) changes at least one parameter (P i ).

16. The method according to claim 1 or 2, wherein the quality criterion describes the degree of freedom from artifacts.

17. The method according to claim 1 or 2, wherein the reference value (R1, ..., Rn) of the test area (Rt) reflects the signal strength.

18. The method according to claim 17, wherein the signal strength is an average signal strength in the test region (Rt). The method of claim 17 , wherein the quality criterion requires the smallest possible signal strength in the test area.

20. The method according to claim 1 or 2, wherein the termination criterion takes into account a preset maximum number of executions of step e) and / or a predetermined maximum number of differences between the reference value (R1, ..., Rn) and the test reference value (R1 i *,…,Rn i *) is the minimum threshold for the results of the comparisons performed.

21. A magnetic resonance apparatus (1), comprising a magnet unit (3), a gradient unit (5), a radio frequency unit (7) and a control device (9), wherein the control device (9) has a radio frequency transmission / reception control device (7') and a parameter modification unit (15), wherein the control device (9) is configured to perform the method according to any one of claims 1 to 20 on the magnetic resonance apparatus (1).

22. A computer program product, comprising a computer program, which can be directly loaded into a memory of a control device (9) of a magnetic resonance device (1), and which has a program structure so that when the computer program is executed in the control device (9) of the magnetic resonance device (1), the steps of the method according to any one of claims 1 to 20 are performed.

23. An electronically readable data carrier having electronically readable control information stored thereon, the electronically readable control information comprising at least one computer program product according to claim 22 and being designed such that, when the data carrier is used in a control device (9) of a magnetic resonance device (1), the electronically readable control information executes a method according to any one of claims 1 to 20.

Images