Correction of interference effects in echoplanar imaging

Abstract

A method for performing magnetic resonance imaging (MRI) with an echo-planar sequence of a region of interest (ROI) is described. In this method, a theoretical model representing an oscillating structure of a magnetic field-generating unit (1b) is generated. Oscillation of this structure induces eddy currents, which in turn generate a gradient-like interference magnetic field in the ROI. This gradient-like interference magnetic field is determined based on the theoretical model (MMF). The echo-planar sequence is then played back. Magnetic resonance signals (RX) from the ROI are received, and raw magnetic resonance data (RD) is generated from these signals.Based on the raw magnetic resonance data (RD) and the determined gradient-like interference magnetic field in the area of ​​investigation (ROI), corrected raw magnetic resonance data (KRD) are generated. Finally, magnetic resonance image data (BD) are reconstructed based on the corrected raw magnetic resonance data (RD). A magnetic resonance imaging system (1) is also described.

Description

The invention relates to a method for carrying out magnetic resonance imaging with an echo planar sequence from an examination region. The invention also relates to a magnetic resonance imaging system with which such a method can be carried out.A magnetic resonance imaging system includes a control device and a scanning unit. The scanning unit has a so-called magnetic field generating unit (referred to in English as "field generating unit"). The magnetic field generation unit comprises a basic field magnet, a gradient coil for generating phase encoding gradients, readout gradients and slice selection gradients, so-called shim coils for improving the homogeneity of the basic magnetic field, and a radio-frequency magnetic coil, also referred to as "body coil". A "body coil", also referred to in the German as a "body coil", can be used for the transmission and / or reception of high-frequency signals.In echo planar sequences (abbreviated as "EPI" sequences), the readout gradients must be switched very quickly to high gradient strengths and switched off again in order to enable a fast scanning of the lines in k-space. In this case, interactions occur between the basic field magnet and the gradient coil, in which oscillations are caused, which can impair the entire measurement. Furthermore, due to the rapid change of the gradients, gradient-induced eddy currents are generated, which arise on electrically conductive structures and in particular also on resonantly excited, electrically conductive structures of the magnetic field generating unit and counteract the build-up of the gradient field. A distinction is made between primary eddy currents which are generated as a result of the magnetic field gradients and secondary eddy currents. Due to the induced primary eddy currents in the basic magnetic field, Lorenz forces are generated. If the Lorenz forces excite mechanical resonances, secondary eddy currents are generated on account of these mechanical oscillations. The present patent application is concerned with the compensation of the secondary eddy currents. This reduction of the gradient field has the effect that the signal maximum is not registered in the center of k-space, but is registered in odd phase coding steps behind the center of k-space and in even phase coding steps before the center of k-space.Artifacts, so-called N / 2 ghosts, arise due to this zigzag line of the signal maxima, in this specific case referred to as "EPI ghosts" or "ghost artifacts". These ghost artifacts are superimposed on the image data with a shift of N / 2 in the phase coding direction. N / 2 corresponds to half the number of phase coding lines which are required for scanning the examination region (also referred to as "region of interest" and abbreviated as "ROI") or the field of view (in the German "field of view", also abbreviated as "FOV") in the phase coding direction. Although the "EPI ghosts" look similar to aliasing artifacts resulting from magnetic resonance signals from tissue outside the FOV generated at a frequency above the Nyquist frequency, they differ from them in that the "EPI ghosts" occur even if the object under investigation is smaller than the FOV in the phase encoding direction and therefore no frequencies above the Nyquist frequency can occur.Similar phenomena also occur in a vibration-induced manner when excited in the mechanical resonance of the gradient coil. The eddy currents can therefore be generated not only directly by rapid magnetic field changes, but also by moving, electrically conductive components through which magnetic fields pass. Compensation methods used to date are not particularly effective or have other adverse effects on imaging sequences used. For example, they extend the total acquisition time and the minimum possible echo time as well as the susceptibility to the magnetic field drift. A magnetic field drift can be caused by heating individual components, in particular the gradient coil or the so-called shim iron.As countermeasures, certain echo times have been prohibited in connection with EPI sequences, in particular echo times corresponding to self-resonances of gradient coils. However, due to the strong attenuation of the oscillation in the magnetic field, such a procedure has the result that a very broad range of echo times must be prohibited. This greatly restricts optimization of imaging sequences.As a further countermeasure, reference recordings were made for phase correction in the steady and / or identical mechanical state of the gradient system. The three phase correction variants described below are known in the prior art.In internal phase correction, also known as "3-line B 0" correction, three reference lines are acquired before each readout phase. The internal phase correction is illustrated in FIG. 1. The time for reading out the reference lines is relatively short, but must be applied in addition to the time for the actual EPI sequence. The readout sequence for the readout of the three reference lines differs from the readout sequence of the actual EPI sequence for image acquisition. The additional time requirement is comparatively low with approximately 5 ms (ms=msec) for each EPI sequence. Overall, internal phase correction is distinguished by short measurement times and a high degree of robustness with respect to patient movements.In the external phase correction, which is illustrated in FIG. 2, during the reading out of the three reference lines, the entire read-out gradient is expanded, which is later also used for the image acquisition of the EPI sequence. In this approach, artifact suppression is particularly strong, but the time required due to the playing out of the entire readout gradient is particularly great when playing out and reading out the reference lines. The additional time requirement is approximately at the repetition time of the EPI sequence. If field deviations occur between different repetitions, the correction can become inaccurate. The external phase correction is also sensitive to patient movements. Otherwise, the image quality is increased by the external phase correction compared to the internal phase correction.In the third known method, which is illustrated in FIG. 3, the reduction of ghost artifacts takes place on the basis of a GRAPPA reference recording. For this purpose, two additional reference recordings are made at the beginning of an EPI sequence in order to record GRAPPA-like kernels with which imperfections of the k-space trajectory scanned during the raw data acquisition are to be corrected. Here too, the complete readout gradient is implemented during the reference recording. The additional time requirement here is twice the repetition time of the actual EPI sequence. If field deviations occur between different repetitions, the correction can become inaccurate.Further descriptions of methods for determining and / or avoiding eddy current-induced artifacts are known, for example, from DE 197 15 113 A1, the doctoral work of Ch. Strohlein, A., "Magnetomechanical interactions in clinical MRI cryostats," 2019, and the article by King et al., "Measurement of Short Time Constant Eddy Currents with Zero TE Imaging," Proc. Intl. Soc. Mag. Resonance. Med. 21: pp. 3777, 2013.It is therefore the object to develop a method for carrying out magnetic resonance imaging with an echo planar sequence with high image quality and increased robustness with respect to movements and with reduced time outlay in comparison with previous procedures.This object is achieved by a method for performing magnetic resonance imaging with an echo planar sequence of an examination region according to claim 1 and a magnetic resonance imaging system according to claim 11.In the method according to the invention for carrying out magnetic resonance imaging with an echo planar sequence from an examination region, a theoretical model is generated on the basis of the echo planar sequence. In an echo planar sequence, a frequency gradient or readout gradient generates a so-called echo train having a plurality of, for example, 128 circuits. This plurality of circuits enables data acquisition of an image in a very short time, for example 70 ms. The echo planar sequence is suitable in particular for recording dynamic physiological processes in the body of a patient. The theoretical model represents an oscillating structure of a magnetic field generating unit, wherein by an oscillation of the oscillating structure, eddy currents are induced in the oscillating structure, which generate a gradient-like disturbance magnetic field in the examination region. On the basis of the theoretical model, a gradient-like interference magnetic field is determined. Furthermore, the echo planar sequence is implemented. As already explained at the beginning, the method according to the invention compensates the secondary eddy currents which are generated on the basis of the mechanical oscillations generated by primary eddy currents. The primary eddy currents also mentioned are preferably compensated by pre-distortion of the gradient fields applied.In response to the echo planar sequence that has been implemented, magnetic resonance signals are received and acquired from the examination region, wherein raw magnetic resonance data are generated on the basis of the magnetic resonance signals. Corrected magnetic resonance raw data are now determined on the basis of the determined gradient-like interference magnetic field in the examination region and on the basis of the magnetic resonance raw data. On the basis of the knowledge of the interference magnetic field, it is possible in particular to carry out corrections of the phase response or of the phase of the magnetic resonance raw data, which corrections lead to a reduction in the N / 2 ghosts described at the beginning.Finally, magnetic resonance image data is reconstructed on the basis of the corrected magnetic resonance raw data.In other words, in the method according to the invention, a theoretical model is first used on the basis of knowledge of the properties of the echo planar sequence and, if appropriate, of system parameters of the magnetic resonance imaging system used for magnetic resonance imaging. The theoretical model describes oscillations generated by primary eddy currents. The oscillations in turn cause Lorenz forces in the magnetic field present, by means of which secondary eddy currents are generated. On the basis of these secondary eddy currents, an interference magnetic field is generated, which is determined on the basis of the theoretical model.Unlike conventional, in the method according to the invention, image correction is carried out on the basis of a theoretical model with which the interference magnetic field is determined. Known system parameters or known system parameter values of the magnetic resonance imaging system assigned to these system parameters can be used to calculate the theoretical model. If necessary, these system parameters can be determined experimentally in advance. In contrast to the conventional approaches, however, these system parameter values need to be determined experimentally only once, if at all, and can then be used as model parameter values if necessary to calculate the interference magnetic field.The magnetic resonance imaging system according to the invention has a model unit for generating a theoretical model, which represents an oscillating structure of a unit generating a magnetic field, on the basis of an echo planar sequence to be played. By oscillations of the oscillating structure, eddy currents, which generate a gradient-like interference magnetic field in an examination region, are induced in the oscillating structure. The model unit also serves to determine the gradient-like interference magnetic field on the basis of the theoretical model.Part of the magnetic resonance imaging system according to the invention is also a sequence unit for implementing the echo planar sequence.The magnetic resonance imaging system according to the invention further comprises an input interface for receiving magnetic resonance signals from the examination region, wherein raw magnetic resonance data are generated on the basis of the magnetic resonance signals.The magnetic resonance imaging system according to the invention comprises a correction unit for ascertaining corrected magnetic resonance raw data on the basis of the magnetic resonance raw data and on the basis of the ascertained gradient-like interference magnetic field in the examination region.The magnetic resonance imaging system according to the invention also comprises a reconstruction unit for reconstructing magnetic resonance image data on the basis of the corrected magnetic resonance raw data. The magnetic resonance imaging system according to the invention shares the advantages of the method according to the invention for carrying out magnetic resonance imaging with an echo planar sequence of an examination region.A large part of the aforementioned components of the magnetic resonance imaging system according to the invention can be realized completely or partially in the form of software modules in a processor of a corresponding computing system, e.g. by a control device of a magnetic resonance imaging system or a computer which is used to control such a system. A realization largely through software has the advantage that even previously used computing systems can be easily retrofitted with a software update in order to operate in the manner according to the invention.In this respect, the object is also achieved by a corresponding computer program product having a computer program which can be loaded directly into a computing system, having program sections in order to carry out the steps of the method according to the invention for carrying out magnetic resonance imaging with an echo planar sequence of an examination region when the program is executed in the computing system. In addition to the computer program, such a computer program product can optionally comprise additional components, such as documentation and / or additional components, also hardware components, such as hardware keys (dongles, etc.) for using the software.For transport to the computing system or to the control device and / or for storage on or in the computing system or the control device, a computer-readable medium, e.g. a memory stick, a hard disk or another transportable or permanently installed data carrier, can be used, on which the program sections of the computer program that can be read in and executed by a computing system are stored. For this purpose, the computing system can have one or more cooperating microprocessors or the like, for example.The dependent claims and the following description each contain particularly advantageous embodiments and further developments of the invention. In this case, in particular, the claims of one claim category can also be developed analogously to the dependent claims of another claim category. In addition, within the scope of the invention, the various features of different exemplary embodiments and claims can also be combined to form new exemplary embodiments.In a preferred embodiment of the method according to the invention for carrying out magnetic resonance imaging with an echo planar sequence from an examination region, the theoretical model comprises a mechanical model which describes the mechanical oscillation behavior of a magnetic field generating unit with which the magnetic resonance imaging is carried out. Advantageously, the oscillations of the magnetic field generating unit caused by the primary eddy currents during the imaging are taken into account in the determination of the secondary eddy currents and the interference magnetic field generated thereby.In a preferred variant of the method according to the invention for carrying out magnetic resonance imaging with an echo planar sequence of an examination region, the determination of the corrected magnetic resonance raw data comprises a phase correction. A deviation of a magnetic field results in a change in the phase of a scanned magnetic resonance signal. To correct this error, the phase of the sampled magnetic resonance signals is corrected based on the theoretical model.Furthermore, the determination of the corrected magnetic resonance raw data preferably comprises a correction of the basic magnetic field B 0. With the change of the basic magnetic field, the Lamber frequency of the excited protons changes and thus also the frequency with which magnetic resonance signals are acquired. By correcting the B 0- field, a shift of the image data in the image data space or spatial space is compensated. Since the pixel bandwidth in the readout direction is different in comparison with the pixel bandwidth in the phase coding direction in an EPI sequence, a distortion of the image data is added in an EPI sequence, which distortion is likewise compensated by the correction of the basic magnetic field B 0.In one embodiment of the method according to the invention for carrying out magnetic resonance imaging with an echo planar sequence from an examination region, the theoretical model is designed such that the eddy currents comprise fast-decaying eddy currents with a time constant between 0.1 and 10 ms. This depends on the material and structural nature of the vibrating structure. Typical values for decay times of eddy currents in oscillating components of a magnetic resonance imaging system lie in the aforementioned value interval. Time constants shorter than 0.1 ms are negligible, since in this case the phase error is too small. Time constants longer than 10 ms would shift all k-space lines and thus cause no N / 2 ghost.Preferably, the theoretical model of the oscillating structure comprises an attenuated harmonic oscillator. The model of the attenuated harmonic oscillator can be completely described by a few parameters, in particular the natural frequency and the attenuation. More complex oscillating structures also behave approximately like a harmonic oscillator in the case of small deflections. Advantageously, the oscillation behavior of a magnetic resonance imaging system can be described in a simple manner in a model form. Furthermore, a wide range of closed analytical solutions is available which describe mechanical oscillation.Preferably, the echo planar sequence comprises a reference measurement for phase change. This reference measurement preferably comprises the internal phase correction mentioned at the beginning, also referred to as "3-line B 0" correction. In this variant, the theoretical model serves to perform a correction of this reference-based internal phase correction in order to further reduce artifacts without increasing the recording time.Preferably, the reference measurement for the phase change, also referred to as a PC scan, comprises a complete, preferably one-time, reference recording of the k-space. By a one-time measurement of the space, k-space line-dependent correction values of the basic magnetic field and of the linear phase response or of the phase can be determined. Since a clinical EPI protocol comprises a plurality of EPI measurements, the total measurement time is hardly extended by a single additional complete measurement of k-space without phase coding.Particularly preferably, the reference measurement takes place in the steady state. If, for example, a complete PC scan is carried out, the internal phase correction, also referred to as "3-line B 0" correction, can be carried out in the steady state, because the resonating structure already continues to resonate during the first half of the PC scan. If, for example, 64 k-space lines are recorded for a PC scan, the vibration is already in the steady state during the recording of the three medium k-space lines as a reference measurement for an external phase correction, but the PC scan is partly in the unconfined state. If the actual image recording is subsequently carried out in the steady state, the internal phase correction can be used as a correct reference measurement because it was likewise carried out in the steady state. If the PC scan is started in the non-steady state, a time saving results.The echo planar sequence likewise particularly preferably comprises an introductory reference measurement for detecting drift effects. In order to determine so-called drift effects, it may be advantageous to perform an additional reference measurement by implementing reference gradients before the imaging and to use their results for correcting system parameter values or for correcting the response behavior of the magnetic resonance imaging system determined by the model. In this advantageous variant, the relatively low time requirement for the reference measurement is combined with a significantly improved image quality on account of the interference field calculated by the model. Advantageously, so-called drift effects are compensated, which can lead to errors, in particular in the case of measurements extended over time.In the method for carrying out magnetic resonance imaging with an echo planar sequence of an examination region, a reference recording is first carried out, which preferably comprises a "3-line B 0 " correction or an internal phase correction. In this recording, a deviation of the basic magnetic field and preferably a deviation of the phase response or a deviation of the phase and at least also a temporal echo-to-echo distance is preferably determined. This temporal echo-to-echo interval indicates the temporal interval between two successive echoes that have been excited by a radio-frequency signal of an echo planar sequence. Due to the described interference effects, this temporal echo-to-echo distance can vary slightly.Subsequently, on the basis of the temporal echo-to-echo distance, a mechanical model of a magnetic field generation unit of a magnetic resonance imaging system with which the imaging is carried out is generated as a theoretical model. The mechanical model represents the mechanical oscillation behavior of the magnetic field generation unit in the form of a harmonic oscillator. On the basis of this model, the contribution of the eddy currents due to the oscillations to generate the basic magnetic field can be estimated in a model-like manner, and the phase response or the phase of the magnetic resonance signal can be calculated as a time delay in the time domain as a function of a k-space line and the excitation frequency.On the basis of this model, an extended correction of the already estimated deviation of the basic magnetic field and the already estimated deviation of the phase response of the magnetic resonance signal is carried out and values corrected in this way for the basic magnetic field and the phase response or the phase are calculated as a function of the ith k space line.Finally, a phase correction of the magnetic resonance raw data and thus also of the image data is carried out on the basis of the corrected field and phase data.In a particularly preferred variant of the method according to the invention for carrying out magnetic resonance imaging with an echo planar sequence of an examination region, a one-time use of a complete phase change scan or a phase change scan recorded in the steady state and a determination of the k-space line-dependent correction values on the basis of the measurement data thus generated additionally take place before the determination of the theoretical model. The generated measurement data were obtained from the examination region. The values thus generated are used as support values for the theoretical model of the magnetic field generation unit.In a particularly preferred variant of the method according to the invention for carrying out magnetic resonance imaging with an echo planar sequence from an examination region, the echo planar sequence comprises a reference measurement for phase change and a temporal echo-to-echo distance is determined on the basis of the reference measurement, the theoretical model being determined on the basis of said temporal echo-to-echo distance. Advantageously, the true true true can be determined.In a preferred embodiment of the method according to the invention for carrying out magnetic resonance imaging with an echo planar sequence from an examination region, the echo planar sequence has a reference measurement for phase change, which has a scanning of at least three reference lines. The reference measurement preferably has exactly three reference lines. Capturing only three reference lines in the readout direction allows saving time for the reference measurement compared to a reference measurement covering the entire k-space of the imaging. Three reference lines in the readout direction make it possible at least to ascertain a piece of coarse information regarding a phase change between the individual reference lines.The invention is explained in more detail below with reference to the attached figures on the basis of exemplary embodiments. The following are shown: FIG. 1 shows a schematic illustration of a pulse sequence diagram of an internal phase correction, FIG. 2 shows a schematic illustration of a pulse sequence diagram of an external phase correction, FIG. 3 shows a schematic illustration of a pulse sequence diagram of a phase correction which takes place on the basis of a GRAPPA reference recording, FIG. 4 is a flow chart illustrating a method for performing magnetic resonance imaging with an echo planar sequence according to an embodiment of the invention, FIG. 5 is a flow chart illustrating a method for performing magnetic resonance imaging with an echo planar sequence according to an alternative embodiment of the invention, FIG. 6 shows a schematic illustration of a magnetic resonance imaging system according to an exemplary embodiment of the invention, FIG. 7 shows a diagram illustrating a measured deviation between two k-space scanning directions as a function of the number of gradient pulses that are output, FIG. 8 shows image representations of a reference image with artifacts which depend on the number of gradient pulses that are output.In FIG. 1, a schematic representation of a pulse sequence diagram of an echo planar sequence with a conventional internal phase correction is illustrated. The pulse sequence diagram shows a series of high-frequency excitation pulses HF in the top row. In the second row, which is marked with "RX", the pulse sequence diagram shows a first readout pulse ADC ref, which represents a reference signal, and a series of readout high-frequency pulses ADC, which represent the magnetic resonance signals of the imaging. In the third row, readout gradients G x are shown both for the reference recording G ref( Grefis also referred to below as a reference gradient) and for the imaging G x( Gxhas here the meaning of a gradient in the readout direction which is output simultaneously with the time window of the acquisition of the readout high-frequency pulse ADC). In the fourth row, phase encoding gradients G y are illustrated. In the fifth row, slice selection gradients G z are shown. As already mentioned, the time required for the actual reference measurement is very short; for this purpose, the reference gradient G ref does not exactly correspond to the gradient G x played out for the imaging in the readout direction.In FIG. 2, a schematic illustration of a pulse sequence diagram of a conventional external phase correction is illustrated. During the external phase correction, a gradient G x is already applied in the readout direction as a reference gradient G ref, during the reference measurement, which gradient corresponds to the gradient G applied simultaneously to the readout high-frequency pulse ADC, in the readout direction. In this way, in the reference measurement, interference signals generated in the readout direction by the gradient G x also played out during the actual imaging are equally detected with the first readout pulse ADC ref generated by a high-frequency reference signal, so that the determination of the interference effects and their compensation is improved in comparison with the time-saving method shown in FIG. 1. However, the external phase correction takes more time than the internal phase correction shown in FIG. 1.FIG. 3 shows a schematic illustration of a pulse sequence diagram of a conventional phase correction which takes place on the basis of a GRAPPA reference recording. In the conventional correction method shown in FIG. 3, two reference gradients G ref are output in the readout direction. The reference measurements or the read-out pulses ADC ref are used for calculating kernels which are included in the image reconstruction and which are intended to compensate for interference effects already occurring in the reference measurements.FIG. 4 shows a flow diagram 400 illustrating a method for carrying out magnetic resonance imaging with an echo planar sequence from an examination region according to an exemplary embodiment of the invention.In step 4.I, a reference recording Int PC SC is initially carried out, which corresponds to the reference measurement of the "3-line B 0" correction or internal phase correction illustrated in FIG. 1. In this recording, in addition to a deviation dB 0 of the basic magnetic field and a change dPH of the phase response or a change of the phase, a temporal echo-to-echo distance T EE is also determined.In step 4.II, a mechanical model MMF of the magnetic field generation unit is generated on the basis of the temporal echo-to-echo distance T EE. The mechanical model MMF represents the mechanical oscillation behavior of the magnetic field generation unit in the form of a harmonic oscillator. Based on this model, the contribution MMF-B 0,i of the eddy currents due to the oscillations to generate the basic magnetic field B 0 can be estimated in a model manner, and an estimated phase response or estimated phase MMF-PH i can be calculated as a time delay in the time domain as a function of a k-space line and the excitation frequency.On the basis of this model, in step 4.III, an extended correction EXT-PC of the deviation dB 0 of the basic magnetic field already determined in step 4.I and the change dPH of the phase response determined in step 4.I is carried out, and values corrected in this way for the basic magnetic field B 0,i and the phase response PH i or the phase are calculated as a function of the i-th k space line.In step 4.IV, a phase correction PC of the magnetic resonance raw data and thus also of the image data is now carried out on the basis of the corrected data B 0,i, PH i.FIG. 5 shows a flow diagram 500 illustrating a method for carrying out magnetic resonance imaging with an echo planar sequence according to an alternative exemplary embodiment of the invention.In the exemplary embodiment illustrated in FIG. 5, step 5.II additionally involves a one-time use of a phase change scan PC-S recorded completely or in the steady state and a determination of the k-space line-dependent correction values B 0i, PH i on the basis of the measurement data generated in this way. The generated measurement data were obtained from the examination region ROI. The values thus generated are used in step 5.III as support values for the mechanical model MMF of the magnetic field generation unit. Steps 5.IV and 5.V correspond to steps 4.III and 4.IV.FIG. 6 shows a schematic illustration of a magnetic resonance imaging system 1 according to an exemplary embodiment of the invention. The magnetic resonance imaging system 1 has a model unit 3 athat is configured to generate a theoretical model MMF representing an oscillating structure of a magnetic field generating unit. In this case, by vibration of the vibrating structure, eddy currents are induced in the vibrating structure, which cause a gradient-like interference magnetic field in the examination region ROI. The model unit 3 ais therefore also configured in particular to determine the gradient-like interference magnetic field on the basis of the vibration of the vibrating structure calculated by model and the eddy currents of the theoretical model resulting therefrom. On the basis of the eddy currents or the resulting interference magnetic field, a change of the basic magnetic field B 0,i and of the phase response PH i of the magnetic resonance signals results.Part of the magnetic resonance imaging system 1 is also a sequence unit 1 afor implementing an echo planar sequence for generating radio-frequency signals RF with which magnetic resonance signals RX are excited. The magnetic resonance imaging system 1 also comprises a magnetic field generation unit 1 b, which generates a basic magnetic field B 0, gradient fields G x, G ref and radio-frequency fields on the basis of the echo planar sequence that has been played out.The magnetic resonance imaging system 1 also comprises an input interface 2 for receiving the magnetic resonance signals RX from the examination region ROI, wherein raw magnetic resonance data RD are generated on the basis of the magnetic resonance signals RX.Part of the magnetic resonance imaging system 1 is also a correction unit 3 for ascertaining corrected magnetic resonance raw data KRD on the basis of the magnetic resonance raw data RD and on the basis of the deviations of the basic magnetic field B 0,i and of the phase response PH i of the magnetic resonance signals RX ascertained by the model unit 3 a.The magnetic resonance imaging system 1 also comprises a reconstruction unit 4 for reconstructing magnetic resonance image data BD on the basis of the corrected magnetic resonance raw data KRD.FIG. 7 illustrates a graph which shows a measured deviation PKPKbetween two k-space scanning directions as a function of the number PZof the bipolar gradient pulses which have been output. The deviation PKPK refers to the deviation of the basic magnetic field in Hz and from peak to peak. In the scenario illustrated in FIG. 7, a phantom is placed about 15 cm out of the isocenter in the x direction and y direction. After a number PZ of 20 bipolar gradient pulses, a transient process is concluded, which is illustrated by the horizontal profile of the PKPK curve at the value PZ=20.FIG. 8 illustrates image representations of a reference recording with artifacts which depend on the number of gradient pulses that are played out. As can be seen in FIG. 8, the artifacts reduce after a number PZ of approximately 10 gradient oscillation periods.Finally, it is pointed out once again that the methods and apparatuses described above are merely preferred exemplary embodiments of the invention and that the invention can be varied by the person skilled in the art without departing from the scope of the invention, insofar as it is specified by the claims. For the sake of completeness, it is also pointed out that the use of the indefinite articles "a" or "an" does not exclude that the features in question can also be present multiple times. Likewise, the term "unit" does not exclude it being composed of a plurality of components which may optionally also be spatially distributed. Regardless of the grammatical sex of a certain term, individuals with male, female or other sex identity are included.

Claims

Method for carrying out magnetic resonance imaging with an echo planar sequence from an examination region (ROI), comprising the steps: - generating a theoretical model (MMF) which represents an oscillating structure of a unit (1b) generating a magnetic field, wherein eddy currents which generate a gradient-like interference magnetic field in the examination region (ROI) are induced in the oscillating structure by an oscillation of the oscillating structure on the basis of the echo planar sequence, - determining the gradient-like interference magnetic field on the basis of the theoretical model (MMF), - implementing the echo planar sequence, - receiving magnetic resonance signals (RX) from the examination region (ROI), wherein raw magnetic resonance data (RD) are generated on the basis of the magnetic resonance signals (RX), determining corrected magnetic resonance raw data (KRD) on the basis of the generated magnetic resonance raw data (RD) and on the basis of the determined gradient-like interference magnetic field in the examination region (ROI), reconstructing magnetic resonance image data (BD) on the basis of the corrected magnetic resonance raw data (KRD), wherein the echo planar sequence comprises a reference measurement for phase change and a temporal echo-to-echo distance (T EE) is determined on the basis of the reference measurement, the theoretical model (MMF) being determined on the basis of said reference measurement.The method according to claim 1, wherein the theoretical model (MMF) comprises a mechanical model describing the mechanical vibration behavior of the magnetic field generating unit (1b) with which the magnetic resonance imaging is performed.Method according to Claim 1 or 2, wherein the determination of the corrected magnetic resonance raw data (KRD) comprises a phase correction.A method according to any preceding claim, wherein the eddy currents comprise rapidly decaying eddy currents having a time constant between 0.1 and 10 ms.Method according to any one of the preceding claims, wherein the theoretical model (MMF) of the oscillating structure comprises an attenuated harmonic oscillator.The method of any preceding claim, wherein the echo planar sequence comprises a phase change reference measurement (PC-S).The method of claim 6, wherein the phase change reference measurement (PC-S) comprises a full reference k-space image.Method according to Claim 6 or 7, wherein the reference measurement for phase change (PC-S) is carried out in the steady state.Method according to any of the preceding claims, wherein the echo planar sequence comprises an preliminary reference measurement for detecting drift effects.The method of any preceding claim, wherein the echo planar sequence comprises a phase change reference measurement having a sample of at least three reference lines.Magnetic resonance imaging system (1), comprising: - a model unit (3a) for generating a theoretical model (MMF) representing an oscillating structure of a unit (1b) generating a magnetic field, wherein by an oscillation of the oscillating structure eddy currents are induced in the oscillating structure, which generate a gradient-like interference magnetic field in the examination region (ROI), on the basis of an echo planar sequence to be applied and for determining the gradient-like interference magnetic field on the basis of the theoretical model (MMF), - a sequence unit (1a) for applying the echo planar sequence, - an input interface (2) for receiving magnetic resonance signals (RX) from the examination region (ROI), wherein magnetic resonance raw data (RD) are generated on the basis of the magnetic resonance signals (RX), - a correction unit (3) for determining corrected magnetic resonance raw data (KRD) on the basis of the magnetic resonance raw data (RD) and on the basis of the determined gradient-like interference magnetic field in the examination region (ROI), - a reconstruction unit (4) for reconstructing magnetic resonance image data (BD) on the basis of the corrected magnetic resonance raw data (KRD), wherein the echo planar sequence comprises a reference measurement for phase change and on the basis of the reference measurement a temporal echo-to-echo distance (T EE) is determined, on the basis of which the theoretical model (MMF) is determined.A computer program product comprising instructions which, when the program is executed by a computer, cause the computer to carry out the steps of the method of any one of claims 1 to 10.A computer readable storage medium comprising instructions which, when executed by a computer, cause the computer to carry out the steps of the method of claims 1 to 10.

Images