Determination of magnetic resonance coil sensitivity data

Abstract

Method for determining magnetic resonance coil sensitivity (SM) data comprising the steps: - Acquiring reference data (RD) of individual layers with a coil comprising at least two antenna elements and one coil channel per antenna element, wherein the acquired reference data (RD) completely sample k-space at least in a central region of k-space according to Nyquist, - Loading at least one Nyquist-undersampled set (SD) of data acquired with the coil, - Determining a supplementary kernel (S*) for at least one loaded set (SD) of data acquired with the coil based on the acquired reference data (RD), - Determining Nyquist-complete, reconstructed reference data (RD*) from the loaded set (SD).the loaded sets (SD) of data recorded with the coil using the determined supplementary kernels (S*),- Determining coil sensitivity data (SM) of the coil based on the reconstructed reference data (RD*), wherein loaded sets (SD) of data recorded with the coil were determined based on the recorded reference data (RD).

Description

The invention relates to an improved determination of magnetic resonance coil sensitivity data.The magnetic resonance technique (hereinafter, the abbreviation MR stands for magnetic resonance) is a known technique with which images of the interior of an examination object can be generated. Expressed in simplified form, for this purpose, the examination object is positioned in a magnetic resonance apparatus in a comparatively strong static, homogeneous basic magnetic field, also called B0field, with field strengths of 0.2 Tesla to 7 Tesla and more, so that its nuclear spins are oriented along the basic magnetic field. To trigger nuclear magnetic resonances measurable as signals, high-frequency excitation pulses (RF pulses) are radiated into the examination object, the triggered nuclear magnetic resonances are measured as so-called k-space data by means of coils designed for reception and MR images are reconstructed or spectroscopy data are determined on the basis thereof. The alternating magnetic field generated by the excitation pulses irradiated by means of a transmitting coil is also referred to as a B1field. For spatial coding of the measurement data, magnetic gradient fields, referred to as gradients for short, which are switched rapidly are superimposed on the basic magnetic field. A scheme used, which describes a temporal sequence of RF pulses to be radiated and gradients to be switched, is referred to as a pulse sequence (scheme), or also briefly as a sequence. The recorded measurement data are digitized and stored as complex numerical values in a k-space matrix. An associated MR image can be reconstructed from the k-space matrix occupied with values, for example, by means of a multidimensional Fourier transformation.In this case, there are basically two ways of generating echo signals after excitation of the nuclear spins. On the one hand, the excited nuclear spins can be manipulated by switching dephaser and rephaser gradients in such a way that the signal decays more quickly than the T2* decay inherent to the measured tissue, but after a certain time, the echo time (TE), a so-called gradient echo (GRE) to be measured forms after the RF excitation pulse used. Such sequences are commonly referred to as GRE sequences. On the other hand, by irradiating at least one RF refocusing pulse after irradiation of an RF excitation pulse after a time, again called the echo time, after the RF excitation pulse, a so-called spin echo (SE) can also be generated, which is measured and the amplitude of which is reduced, however, according to the T2decomposition inherent to the measured tissue. Such sequences are commonly referred to as SE sequences. In any case, the excitation and the measurement of the generated echo signals are repeated for each sequence, if appropriate (for example, while switching different gradients for spatial coding) until the desired number of echo signals has been measured and stored in k-space in order to be able to image the examination object.Among the SE sequences, in particular the TSE sequences (TSE: "Turbo Spin Echo"), which are also known under the names FSE ("Fast Spin Echo") or RARE ("Rapid Acquisition with Referenced Echoes") sequences, are widely used in clinical application. The advantage of the TSE sequences compared to the "simple" SE sequence is that a plurality of refocusing pulses are switched after an RF excitation pulse, and that a plurality of spin echo signals are also generated as a result. This accelerates data recording.In so-called "single shot" methods, the entire k-space data to be recorded, e.g. for imaging a slice to be imaged of an examination object, can be recorded after only one RF excitation by an RF excitation pulse.An example of such a single-shot TSE sequence is the HASTE sequence ("half-Fourier acquisition single-shot turbo spin echo imaging"), in which a "partial Fourier" method, in particular the half-Fourier method, is additionally used for reducing the k-space data to be recorded. In this case, the symmetry of k-space with respect to complex conjugation is used in order to derive nonmeasured k-space data from the measured k-space data. Thus, after only one excitation pulse, all k-space data of a slice to be imaged that are necessary for the method can be recorded. If slices of an examination object are to be measured, all required k-space data of a slice can be recorded by means of HASTE after only one excitation. Therefore, HASTE techniques are commonly used for thorax or abdominal imaging where they allow coverage of relatively large volume of interest (VOI) within one or more breath hold phases with reduced sensitivity to physiological movements of the object under investigation.HASTE acquisition techniques are also known, inter alia, among the acronyms SS-FSE (single-shot fast spin echo), SSH-TSE (single-shot turbo spin echo), UFSE (ultra-fast spin echo), single-shot fast SE or FASE or super-FASE (fast advanced spin echo).In addition to the above-mentioned partial Fourier methods, so-called parallel acquisition techniques (e.g. GRAPPA ("GeneR Autocalibrating Partially Parallel Acquisition") and SENSE ("SENSITIVITY ENCODING") ARE ALSO KNOWN, WITH THE AID OF WHICH ACQUISITION TIMES REQUIRED FOR RECORDING THE DESIRED DATA CAN BE SHORTENED, AND WHICH ARE THEREFORE ALSO REFERRED TO AS ACCELERATION TECHNIQUES. As in the case of partial Fourier methods, only parts of the echo signals actually to be recorded as k-space data according to the Nyquist condition are measured in this case. In contrast to the partial Fourier methods, however, the non-measured parts are usually distributed more uniformly over the k-space to be measured according to Nyquist in parallel acquisition techniques, so that, for example, only every second k-space line is measured. In addition, the "missing" k-space data in parallel acquisition techniques are reconstructed with the aid of coil sensitivity data which characterize the coils used for measuring the measurement data, which in this case consist of a plurality of antenna elements each having an associated coil channel. These coil sensitivity data of the coils used in acquiring the measurement data are obtained from reference data that completely scan at least a region of the k-space to be measured, usually the central region, according to the Nyquist condition, and may also be referred to as sensitivity maps of the respective coils. The coil sensitivity data determined from the reference measurement data for a reconstruction of missing k-space data are specific for the respective subsample of the measurement data defining the "missing" and therefore to be added k-space data and are also referred to in this context as supplementary kernel, or for short: "kernel", in the case of GRAPPA also as "GRAPPA kernel".The desire for increasingly faster MR recordings in the clinical environment currently leads to a renaissance of further acceleration techniques, in particular methods in which a plurality of images are recorded simultaneously. In general, these methods can be characterized in that transverse magnetization of at least two slices is used simultaneously for the imaging process in a targeted manner during at least part of the measurement ("multi-slice imaging", "slice multiplexing"). In contrast, in the established "multi-slice imaging", the signal of at least two slices is recorded alternately, i.e. completely independently of one another, with a correspondingly longer measurement time.Known methods for this are, for example, what is known as Hadamard coding, methods with simultaneous echo refocusing, methods with broadband data recording or also methods which use parallel imaging in the slice direction. The latter methods also include, for example, the CAIPIRINHA technique as described by Breuer et al. in "Controlled Aliasing in Parallel Imaging Results in Higher Acceleration (CAIPIRINHA) for Multi-slice Imaging", Magnetic Resonance in Medicine 53, 2005, pp. 684-691, and the buffed CAIPIRINHA technique as described by Setsompop et al. in "Buffed-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.In particular in the latter slice multiplexing methods, a so-called multi-band RF pulse is used to excite or otherwise manipulate two or more slices simultaneously (simultaneously), e.g. to refocusing or saturate. Such a multi-band RF pulse is, for example, a multiplex of individual RF pulses which would be used for manipulation of the individual layers to be simultaneously manipulated. By multiplexing, for example, a baseband-modulated multi-band RF pulse is obtained from an addition of the pulse shapes of the individual RF pulses. The spatial coding of the recorded signals is achieved essentially by a conventional gradient switching in two directions (two-dimensional gradient coding). The multi-band RF pulse is spatially selectively constrained in effect to the desired layers.The signals produced from all excited slices are collapsed into a data set by means of a plurality of receiving antennas and then separated after the individual slices with the aid of parallel acquisition techniques.As already mentioned above, parallel acquisition techniques can already generally shorten the acquisition time required for recording the desired data by a Nyquist incomplete acquisition time, i.e. sub-sampling of the k-space.In slice multiplexing methods, parallel acquisition techniques are used to separate the measurement data recorded simultaneously for different slices again with the aid of coil sensitivity data of the coils used. Reference data must be recorded for all the slices concerned. This is generally carried out within the scope of an additionally to be carried out reference measurement which measures reference data individually for each desired slice, from which coil sensitivity data can be determined.A further acceleration technique is the so-called compressed sensing technique, in which the k-space is likewise undersampled, but this is done with the most (pseudo) random possible distribution of the measured k-space points. Compressed sensing (CS) is described in more detail, for example, in Lustig et al. "Sparse MRI: The application of compressed sensing for rapid MR imaging." Magnetic Resonance in Medicine: An Official Journal of the International Society for Magnetic Resonance in Medicine 58.6 (2007): 1182-1195 or also in Lustig et al. "Compressed sensing MRI." IEEE signal processing magazine 25.2 (2008): 72-82. Here, data on non-measured k-space points, "missing" measurement data, are determined by means of iterative reconstruction functions with addition of a priori knowledge, which enter into the optimization problem of the reconstruction function as boundary conditions and can likewise comprise coil sensitivity data.For some years, it has been further known to use trained reconstruction functions which comprise neural networks based on so-called deep learning (DL) algorithms in order to reconstruct image data from measurement data picked up undersampled by means of magnetic resonance technology. Coil sensitivity data are also used within the scope of such trained reconstruction functions, for example, in order to ensure data consistency.An overview of the principles of parallel acquisition techniques, CS and associated iterative reconstruction functions up to trained reconstruction functions is described in the article by Knoll et al., "Deep-Learning Methods for Parallel Magnetic Resonance Imaging Reconstruction: A Survey of the Current Approaches, Trends, and Issues", IEEE Signal Processing Mag. 37(1): pp. 128-140 (2020).For the aforementioned techniques, which use coil sensitivity data in the reconstruction of image data from the measured measurement data, the accuracy and quality of the coil sensitivity data used is decisive for a successful image reconstruction. Disturbing influences, e.g. due to movement or pulsation during a recording of reference data, on the basis of which coil sensitivity data are determined, can therefore lead to artifacts in image data reconstructed using the coil sensitivity data determined in this way. Particularly susceptible are measurements that use a TSE sequence to record the measurement data for imaging and separate reference measurements to record the reference data. In such measurements, for example, reference data fully sampled within a few seconds are initially recorded for all N slices to be recorded, for example by means of HASTE reference measurements. The acquisition of the measurement data for the imaging then usually takes place, wherein the measurement data is acquired in an accelerated manner, i.e. undersampled manner. For example, in m echo trains, parts, e.g. k-space lines, of k-space are recorded for all N slices. If reference data were recorded under a disturbance, for example in a phase of pulse-like flow, then usually large parts of the k-space, where reference data were recorded, are contaminated by the disturbance. As a result of the distributed recording of the measurement data for imaging over a plurality of time periods, in many cases only individual k-space trajectories, e.g. individual k-space lines, are affected there.The invention is based on the object of making possible an improved determination of coil sensitivity data which in particular reduces contamination of the reference data by disturbing influences.The object is achieved by a method for determining magnetic resonance coil sensitivity data according to claim 1, a magnetic resonance system according to claim 12, a computer program according to claim 13, and an electronically readable data carrier according to claim 14.A method according to the invention for determining magnetic resonance coil sensitivity data comprises the following steps:a coil which comprises at least two antenna elements and one coil channel per antenna element, recording reference data of individual layers, the recorded reference data completely scanning the k-space at least in a central region of the k-space according to Nyquist,loading at least one Nyquist subsampled set of data collected with the coil,determining a supplementary kernel for at least one loaded set of data recorded with the coil on the basis of the recorded reference data,determining, according to Nyquist, complete reconstructed reference data from the loaded sets of data recorded with the coil at least in a central region of the k-space using the determined supplemental kernel,determining coil sensitivity data of the coil on the basis of the reconstructed reference data.By ascertaining reconstructed reference data according to the invention on the basis of which the desired coil sensitivity data are determined, interference effects in recorded reference data can be relativeized, as a result of which susceptibility to artifacts in image data reconstructed using the determined coil sensitivity data is reduced and an achievable image quality is thus increased.By reconstructing the reconstructed reference data, which can be carried out, for example, by means of a GRAPPA technique, disturbing influences, for example due to pulsations during the acquisition of the reference data, are reduced, since the supplementary kernels are obtained by averaging over the entire k-space, and thus entail a globally optimal solution. Thus, a portion of local disturbances is reduced. Moreover, disturbances that exist further often have a different phase in the artificially generated k-spaces and thus advantageously average out.A magnetic resonance system according to the invention comprises a magnetic unit, a gradient unit, a radio-frequency unit and a control device with a reconstruction unit designed to carry out a method according to the invention.A computer program according to the invention implements a method according to the invention on a control device when it is executed on the control device. For example, the computer program comprises instructions which, when the program is executed by a control device, e.g. a control device of a magnetic resonance system, cause this control device to execute a method according to the invention. The control device can be designed in the form of a computer.The computer program can also be present in this case in the form of a computer program product which can be loaded directly into a memory of a control device, having program code means in order to carry out a method according to the invention when the computer program product is executed in a computing unit of the computing system.A computer-readable storage medium according to the invention comprises instructions which, when executed by a control device, for example a control device of a magnetic resonance system, cause the latter to execute a method according to the invention.The computer-readable storage medium can be configured as an electronically readable data carrier which comprises electronically readable control information stored thereon, which comprises at least one computer program according to the invention and are configured such that, when the data carrier is used in a control device of a magnetic resonance system, they carry out a method according to the invention.The advantages and explanations given with respect to the method also apply analogously to the magnetic resonance system, the computer program product and the electronically readable data carrier.Further advantages and details of the present invention are evident from the exemplary embodiments described below and on the basis of the drawings. The examples listed do not constitute a limitation of the invention. The following are shown: FIG. 1 shows a schematic flow diagram of a method according to the invention for determining coil sensitivity data, FIG. 2 shows a schematic diagram of a possible sequence of a determination of reconstructed reference data RD*, FIG. 3 shows a schematic diagram of a further possible sequence of determining reconstructed reference data RD*, FIG. 4 shows a schematically illustrated magnetic resonance system according to the invention.FIG. 1 is a schematic flow diagram of a method according to the invention for determining (magnetic resonance) coil sensitivity data SM.With a coil comprising at least two antenna elements and one coil channel per antenna element, reference data RD of individual layers are recorded (block 101), wherein the recorded reference data RD completely sample the k-space at least in a central region of the k-space according to Nyquist. The acquisition of the reference data RD can take place in a conventional manner, for example by means of separate HASTE reference measurements.At least one Nyquist subsampled set SD of data collected by the coil is loaded (block 103).For at least one loaded set SD of data recorded with the coil, a supplemental kernel S* is determined on the basis of the recorded reference data RD, which is adapted to the respective subsample of the loaded set SD of data recorded with the coil (block 105). This can be done in the usual manner.Using the determined supplemental kernel S*, at least in a central region of k-space according to Nyquist, full reconstructed reference data RD* is obtained from the loaded sets SD of data collected with the coil (block 107). The determination of the reconstructed reference data RD* can supplement the incomplete loaded sets SD with the coil of recorded data to form complete sets of data using the supplementary kernel S*.FIG. 2 shows a possible sequence of a determination of reconstructed reference data RD*. In the case shown, reference data RD recorded along k-space lines running in the readout direction kx and spaced apart in the phase coding direction ky are divided into two sets of subsampled reference data RD1' and RD2' (block 107.1), for example by allocating every second k-space line to the second subsampled set RD2' of reference data and the other k-space lines remaining in a first, then also subsampled set RD1' of reference data.A loaded set SD of data recorded with the coil can thus have been determined on the basis of the recorded reference data RD. The reference data RD is data recorded with the desired coil and is present. The reference data RD is entirely in a central region of the k space according to Nyquist. In order to obtain subsampled sets SD of data captured with the coil based on the captured reference data RD, the fully captured reference data RD may be divided into sets of subsampled data loaded as sets SD of data captured with the coil. Generally, when the reference data RD is acquired in k-space lines, n sets SD of acquired data can be easily generated by associating each nth k-space line of the reference data RD with one set SD of acquired data.At least one or also all of the subsampled sets RD1' and RD2' thus generated from the recorded reference data RD can be supplemented in each case using the associated supplementary kernel S* to form sets of data RD1" and RD2" that are complete according to Nyquist (block 107.2), for example according to a GRAPPA technique. If the recorded reference data RD has been divided into more than two sets of subsampled reference data RD1', RD2', the procedure can be analogous.If at least two sets SD of recorded data have been loaded, which have each been supplemented using the associated supplementary kernel S* to form sets of data RD1", RD2" that are complete according to Nyquist, the reconstructed reference data RD* can be ascertained on the basis of the at least two supplemented sets RD1", RD2" of data (block 107.3).The determination of the reconstructed reference data RD* can here comprise, for example, averaging of supplemented sets RD1", RD2" of data.It is also conceivable that the determination of the reconstructed reference data RD* comprises a discarding of non-supplemented data in the supplemented sets RD1", RD2" of data, so that no measured reference data are included any longer in the reconstructed reference data RD*, but the reconstructed reference data RD* consist purely of reconstructed data.Additionally or alternatively, the determination of the reconstructed reference data RD* can comprise a comparison of at least two supplemented sets RD 1" and RD 2" of data, on the basis of which contaminated parts of the k-space of the supplemented sets RD 1" RD 2" of data can be identified and, if appropriate, discarded. Such a comparison can be made, for example, by simple subtraction of the k spaces of RD1" and RD2" and subsequent summing in the kx direction. This allows individual k-space lines which are severely affected by disturbances to be identified and discarded. Ideally, both k spaces would be largely identical. The supplemented sets RD1" and RD2" of data generally comprise sufficient k-space lines that discarding individual ones of these k-space lines does not have a great influence.If measurement data MD have been recorded (block 100), e.g. accelerated and thus undersampled, from which image data BD are to be reconstructed using coil sensitivity data, a loaded set SD of data recorded with the coil may also have been determined here on the basis of the undersampled measurement data MD. In particular, undersampled measurement data MD measured for imaging can be trimmed to its region central in k-space and this central part MDc of the measurement data MD loaded as a set SD of measured data.FIG. 3 shows a possible sequence for a determination of reconstructed reference data RD* for such an example.In the case shown here, a parallel acquisition technique is used to speed up, and measurement data MD recorded undersampled therewith are cut to their central region (block 107.1'), so that only the k-space lines central in k-space can be loaded as data MDc.A set of cropped measurement data MDc loaded as a set SD of data recorded with the coil can be augmented using the associated augmentation kernel S* to Nyquist-complete reconstructed reference data RD* (block 107.2'), for example according to a GRAPPA technique.On the basis of the reconstructed reference data RD*, coil sensitivity data SM of the coil are determined (block 109), which can be used for a reconstruction of image data BD from measurement data MD (block 110), e.g. within the scope of a CS method, a parallel acquisition method, or a reconstruction function trained within the scope of a trained reconstruction function. The coil sensitivity data SM can be determined, for example, with the aid of an ESPIRiT method on the basis of the reconstructed reference data, as is described, for example, in Uecker et al. "ESPIRiT-An Eigenvalue Approach to Autocalibrating Parallel MRI: Wheel SENSE Meets GRAPPA", Magn. Resonance. Med. 71:990-1001, 2014.FIG. 4 schematically illustrates a magnetic resonance system 1 according to the invention. This comprises a magnetic unit 3 for generating the basic magnetic field, a gradient unit 5 for generating the gradient fields, a high-frequency unit 7 for irradiating and receiving high-frequency signals, and a control device 9 designed to carry out a method according to the invention.In FIG. 4, these subunits of the magnetic resonance system 1 are only shown roughly schematically. The high-frequency unit 7 can consist of a plurality of subunits and comprise, for example, a plurality of coils. In particular, the radio-frequency unit 7 can comprise a body coil which is fixedly integrated in the magnetic resonance system 1 and can in turn comprise, for example, two antenna elements 7.1 and 7.2. Furthermore, the high-frequency unit 7 can comprise one or more different local coils 7*, which can be configured either only for transmitting high-frequency signals or only for receiving the triggered high-frequency signals or for both, and can in turn comprise a plurality of antenna elements and associated coil channels.For the examination of an examination object U, for example a patient or even a phantom, the examination object U can be introduced on a couch L into the magnetic resonance system 1 in its measurement volume. The slices S 1 or S 2 represent exemplary target volumes of the examination object from which echo signals can be recorded and acquired as measurement data.The control device 9 serves for controlling the magnetic resonance system 1 and can control, in particular, the gradient unit 5 by means of a gradient controller 5' and the radio-frequency unit 7 by means of a radio-frequency transmission / reception controller 7'. The high-frequency unit 7 can comprise a plurality of channels on which signals can be transmitted or received.The high-frequency unit 7 together with its high-frequency transmission / reception control 7' is responsible for generating and radiating (transmitting) an alternating high-frequency field for manipulating the spins in a region to be manipulated (for example in slices S to be measured) of the examination object U. In this case, the center frequency of the high-frequency alternating field, also referred to as a B1 field, is generally set as far as possible in such a way that it is close to the resonant frequency of the spins to be manipulated. Deviations from the center frequency from the resonant frequency are referred to as off-resonance. To generate the B1 field, currents controlled by means of the radio-frequency transmission / reception controller 7' are applied to the RF coils in the radio-frequency unit 7.Furthermore, the control device 9 comprises a reconstruction unit 15 for determining reconstructed reference data according to the invention. The control device 9 is configured overall to carry out a method according to the invention.A computing unit 13 comprised by the control device 9 is designed to carry out all the computing operations necessary for the necessary measurements and determinations. Intermediate results and results required for this or determined in this case can be stored in a memory unit S of the control device 9. The units shown are not necessarily to be understood as physically separate units, but merely represent a subdivision into units of meaning, which can also be realized, for example, in fewer or else in just a single physical unit.Via an input / output device I / O of the magnetic resonance system 1, control commands can be directed to the magnetic resonance system, e.g. by a user, and / or results of the control device 9, such as image data, can be displayed.A method described herein can also be present in the form of a computer program which comprises instructions which execute the described method on a control device 9. A computer-readable storage medium may likewise be present which comprises instructions which, when executed by a control device 9 of a magnetic resonance system 1, cause the latter to execute the described method.Regardless of the grammatical sex of a certain term, individuals with male, female or other sex identity are included.References included in the specificationThis list of documents cited by the applicant has been produced in an automated manner and is only included for the better information of the reader. The list is not part of the German patent application or utility model application. The DPMA does not take any adhesion for any faults or omissions.Cited Non-Patent LiteratureControlled aliasing in parallel imaging results in higher acceleration (CAIPIRINHA) for multi-slice imaging", magnetic resonance in medicine 53, 2005, pp. 684-691

[0010] Setsompop et al., "Flipped-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

[0010] Lustig et al., "Sparse MRI: The application of compressed sensing for rapid MR imaging." Magnetic Resonance in Medicine: An Official Journal of the International Society for Magnetic Resonance in Medicine 58.6 (2007): 1182-1195

[0015] Lustig et al., "Compressed sensing MRI." IEEE signal processing magazine 25.2 (2008): 72-82

[0015] Knoll et al., "Deep-Learning Methods for Parallel Magnetic Resonance Imaging Reconstruction: A Survey of the Current Approaches, Trends, and Issues", IEEE Signal Processing Mag. 37(1): pp. 128-140 (2020

[0017] Uecker et al., "ESPIRiT-An Eigenvalue Approach to Autocalibrating Parallel MRI: When SENSE Meets GRAPPA", Magn. Resonance. Med. 71:990-1001, 2014

[0047]

Claims

Method for determining magnetic resonance coil sensitivity data (SM) comprising the steps: - With a coil comprising at least two antenna elements and one coil channel per antenna element, recording reference data (RD) of individual slices, wherein the recorded reference data (RD) completely sample the k-space according to Nyquist at least in a central region of the k-space, - Loading at least one set (SD) undersampled according to Nyquist of data recorded with the coil, - Determining a supplementary kernel (S*) for at least one loaded set (SD) of data recorded with the coil on the basis of the recorded reference data (RD), - Determining data complete at least in a central region of the k-space according to Nyquist, reconstructed reference data (RD*) from the loaded sets (SD) of data recorded with the coil using the determined supplemental kernel (S*), determining coil sensitivity data (SM) of the coil based on the reconstructed reference data (RD*).The method of claim 1, wherein loaded sets of data captured with the coil scan the central area of k-space.Method according to one of the preceding claims, wherein at least two sets of recorded data are loaded, which are each supplemented using the associated supplementary kernel to form sets of data which are complete according to Nyquist, and the reconstructed reference data is ascertained on the basis of the supplemented sets of data.The method of claim 3, wherein the determination of the reconstructed reference data comprises averaging supplemented sets of data.Method according to any of the preceding claims, wherein loaded sets of data recorded with the coil have been determined on the basis of the recorded reference data.The method of claim 5, wherein the fully acquired reference data is divided into sets of subsampled data loaded as sets of data acquired with the coil.Method according to one of claims 5 or 6, wherein the reference data is recorded in k-space lines and n sets of recorded data are generated by associating each nth k-space line with a set of recorded data.The method of any of claims 3 to 7, wherein determining the reconstructed reference data comprises discarding uncomplemented data in the complemented sets of data.Method according to one of claims 3 to 8, wherein the determination of the reconstructed reference data comprises a comparison of at least two supplemented sets of data, on the basis of which contaminated parts of the k-space of the supplemented sets of data are identified and, if appropriate, discarded.Method according to one of claims 1 to 4, wherein loaded sets of data recorded with the coil have been determined on the basis of recorded measurement data, from which image data is also to be reconstructed.The method of claim 10, wherein subsampled measurement data measured for imaging is cropped to its central region in k-space and then loaded as a set of measured data.Magnetic resonance system (1) comprising a magnetic unit (3), a gradient unit (5), a radio-frequency unit (7) and a control device (9) with a radio-frequency transmission / reception controller (7') and with a reconstruction unit (15), wherein the control device (9) is designed to execute a method according to one of Claims 1 to 11 on the magnetic resonance system (1).A computer program comprising instructions which, when the program is executed by a control device (9) of a magnetic resonance system (1), cause the latter to execute the method according to one of claims 1 to 11.A computer readable storage medium comprising instructions which, when executed by a controller (9) of a magnetic resonance apparatus (1), cause the same to carry out the method according to any one of claims 1 to 11.

Images