Method, device and system for correcting the effect of a disturbance on a gradient system

Description

[0001] The invention relates to a method for correcting the effect of an interference effect on a gradient system of a magnetic resonance device during a magnetic resonance measurement. Furthermore, the invention relates to a magnetic resonance device and a magnetic resonance system with a correction device.

[0002] In a magnetic resonance system, a specimen is typically exposed to a magnetic field with a magnetic field strength between 0.5 Tesla and 7 Tesla using a magnetic field system. When the magnetic field is applied, atomic nuclei in the specimen align themselves along the magnetic field with a non-vanishing nuclear magnetic dipole moment, often referred to as spin. Such collective spin behavior is also referred to as "magnetization" on a macroscopic level. Magnetization is the vector sum of all microscopic magnetic moments in the specimen at a specific location.

[0003] In addition to the basic magnetic field, a gradient system is used to apply a magnetic field gradient, which determines the magnetic resonance frequency (Larmor frequency) at the respective location. High-frequency excitation signals or high-frequency fields (RF pulses) are then emitted via a high-frequency transmitter unit using suitable antenna devices. This is intended to cause the spins of certain atomic nuclei resonantly excited by this high-frequency field (i.e., at the Larmor frequency present at the respective location) to be flipped by a defined flip angle relative to the magnetic field lines of the basic magnetic field. If such an RF pulse acts on spins that are already excited, they can be flipped to a different angular position or even flipped back to an initial state parallel to the basic magnetic field. When the excited spins relax, high-frequency signals, so-called magnetic resonance signals, are emitted resonantly.These magnetic resonance signals can be received using suitable receiving antennas (antenna devices, also called magnetic resonance coils or receiving coils) and then demodulated, digitized, and further processed as so-called "raw data." The magnetic resonance signals are received in a spatial frequency domain, known as "k-space," where, during a magnetic resonance measurement,... z.For example, a slice of k-space is traversed temporally along a "gradient trajectory" (also called a "k-space trajectory") defined by the switching of the gradient pulses. The RF pulses are emitted in a timely and coordinated manner. From the acquired raw data, image data of the object under examination can then be reconstructed using a two-dimensional Fourier transform. Alternatively, three-dimensional volumes can now also be defined, excited, and read out, with the raw data then being sorted into a three-dimensional k-space after further processing steps. Reconstruction of three-dimensional image data can accordingly be performed using a three-dimensional Fourier transform.

[0004] Typically, specific sequences of RF pulses and gradient pulses, so-called pulse sequences, are used to control a magnetic resonance system during a magnetic resonance measurement. The pulse sequences can be played in different spatial directions and / or readout time windows, during which the receiving antennas are switched to receive mode and receive the magnetic resonance signals.

[0005] The gradient pulses are typically defined by a gradient amplitude, a gradient pulse duration, and a slope or the first derivative of the pulse shape dG / dt of the gradient pulses ("slew rate"). Since the gradient system has a maximum load limit, the strength and slew rate of the gradient pulses are fundamentally limited.

[0006] Using a so-called measurement protocol, the aforementioned pulse sequences are pre-parameterized for a desired examination, for example, a specific contrast of the calculated images. The measurement protocol can also contain additional control data for the magnetic resonance measurement. In principle, there are a variety of ways in which the pulse sequences can be structured to obtain the desired image data of the examination subject.

[0007] In the following, "magnetic resonance images" are understood to mean image data of the interior of an examination subject generated using a magnetic resonance device controlled within the scope of the method, but also parameter maps that represent a spatial or temporal distribution of specific parameter values ​​within the examination subject and can be generated, for example, from the image data. An "acquisition" of magnetic resonance image data is understood to mean the performance of a magnetic resonance measurement using a magnetic resonance system.

[0008] As described above, magnetic resonance measurements first involve measuring magnetic resonance signals, whose amplitudes are interpreted as Fourier transforms of the image data in k-space. K-space can be understood as a spatial frequency space of a density distribution of the magnetic moments in a region under investigation, in which magnetic resonance signals are acquired. If k-space is sampled with sufficient precision, the spatial distribution of the density of the magnetic moments can be obtained using a Fourier transform (two-dimensional in the case of layer-by-layer scanning). Typically, k-space is scanned line-by-line along a Cartesian grid. However, other scanning patterns are also conceivable.

[0009] In magnetic resonance measurements, the accuracy of the gradient pulse delivery by the gradient system has a strong influence on the quality of acquired image data. In addition to pulse sequences with Cartesian gradient pulse trajectories, pulse sequences with radial or spiral trajectories can also be used. Single-shot EPI sequences are also used for magnetic resonance measurements. The latter sequences place higher demands on the temporal accuracy of the magnetic field gradients. Causes for deviations between the delivered gradient pulses and the expected gradient pulses include, for example, the occurrence of eddy currents, timing and gain errors, and field fluctuations caused by mechanical vibrations after switching magnetic field gradients, as well as thermal variations and / or non-linear behavior of hardware components, such as the memory chips.an amplifier and / or gradient coil of the gradient system. Such interference effects lead to gradient profiles that deviate from the expected gradient profiles. This deviation leads to errors in the determination of the k-space trajectory, errors in the acquired signal, and ultimately to artifacts in the image data.

[0010] If the deviations are precisely known, an actual k-space trajectory can be determined and used for image reconstruction. Alternatively, the actual k-space trajectory can also be measured. Typically, gradient pulses are corrected for a specific gradient system state at a single temperature. For example, a gradient system transfer function is determined once for room temperature and then used for the entire imaging sequence, without considering temperature changes during imaging.

[0011] The one-time, static gradient correction is based on a technique that uses a transfer function, such as a gradient impulse response function (GIRF "Gradient Impulse Response Function" ) or a gradient system transfer function (GSTF "Gradient System Transfer Function" ) according to equation (1) to correct deviations. Where F is the Fourier transform, f is the frequency, and t is the time. GSTF f = F GIRF t

[0012] This technique utilizes linear and time-invariant characteristics of the dynamic gradient system to correct gradient pulses during image reconstruction. One technique uses the gradient impulse response function (GIRF) or the gradient system transfer function (GSTF) of the magnetic resonance system to correct non-Cartesian trajectories. Using the gradient system transfer function (GSTF) allows the behavior of the entire gradient system to be characterized.

[0013] In a so-called post-correction, corrected gradient pulses g post,l (t) are calculated for each axis (l = x, y, z) by multiplying the Fourier transform of the nominal gradient G nom,l (f) with corresponding gradient system transfer functions at a specific temperature state GSTF l,l (f) and a Fourier transformation into the time domain (back transformation): g post , l t = F − 1 F g nom , l t ⋅ GSTF l , l f

[0014] The value "l" represents the direction x, y, z in which an input gradient g nom,l (t), also called nominal gradient, was played out and in which an output gradient G real,l (f) is measured.

[0015] Alternatively, the inverse gradient impulse response function or gradient system transfer function is used for a preliminary equalization (so-called "pre-emphasis"). Here, a corrected, nominal gradient pulse g pre,l (t) can be output, which corresponds to the desired, nominal gradient pulse g nom,l , so that a desired output gradient G real,l (t) is achieved: g pre , l t = F − 1 F g nom , l t ⋅ GSTF l , l − 1 f

[0016] Equations 2 and 3 show simplified corrections of the gradient pulses, using only first-order self-terms of the gradient system transfer function. For clarification, it should be briefly mentioned here that the gradient system transfer function also includes terms that describe interactions between magnetic field components of different directions, as well as higher-order terms. To perform a comprehensive and complete correction, all terms of the gradient system transfer function from the 0th to the nth order must be used. However, a sufficient correction can already be achieved by primarily using the 0th-order terms to correct the fundamental magnetic fields (B 0 fields) and using the first-order terms to correct the magnetic field gradients.When determining the gradient system transfer function, the gradient system should be tested over a broad frequency spectrum. Ideally, a Dirac pulse is used as the test gradient function to ensure coverage of all frequencies.

[0017] The gradient pulses generated by the gradient system of the magnetic resonance scanner are typically measured in small dynamic field probes, which can be used to determine a phase evolution or phase signal. The phase signal is based on a raw signal, i.e., the signal measured by the receiving antennas. Only the phase of the raw signal is further processed. The magnitude is ignored. The gradient can be determined from the phase signal.

[0018] The correction of gradient pulses is thus already known. However, existing techniques are limited to the correction of k-space data after the acquisition of magnetic resonance signals or to the correction of interference effects that constantly occur during a magnetic resonance acquisition. Compensation for dynamic, particularly nonlinear, behavior of components of the gradient system of the magnetic resonance device during magnetic resonance measurements has not yet been considered.

[0019] Nussbaum et al. propose dividing a gradient chain into a cascade consisting of eddy current compensation, amplifier, and coil. The influences of the amplifier and coil can be separated using field measurements through current measurement, and linearity and time-variance violations are analyzed through heating experiments and varying test pulses (Nussbaum Jennifer et al.: "Nonlinearity and thermal effects in gradient chains: a cascade analysis based on current and field sensing", PROCEEDINGS OF THE INTERNATIONAL SOCIETY FOR MAGNETIC RESONANCE IN MEDICINE, ISMRM, 27TH ANNUAL MEETING AND EXHIBITION, MONTREAL, QUEBEC, CANADA, 11 MAY - 16 MAY 2019, No. 213, April 26, 2019 [2019-04-26], XP040707600).

[0020] Document EP 3 543 722 A1 discloses a magnetic resonance system with a gradient coil system and a current sensor system configured to measure current sensor data describing electrical currents supplied to gradient coils of the gradient coil system. A corrected k-space trajectory is calculated using the current sensor data and a gradient coil transfer function, and a corrected magnetic resonance image is reconstructed using magnetic resonance image data and the corrected k-space trajectory.

[0021] Rahmer et al. determine precise field dynamics by measuring gradient amplifier output currents concurrently with imaging. The actual dynamic field evolution is determined by convolution with a measured current-to-field impulse response of a gradient coil (Rahmer Jürgen et al.: "Non-Cartesian k-space trajectory calculation based on concurrent reading of the gradient amplifiers' output currents", MAGNETIC RESONANCE IN MEDICINE, Vol. 85, No. 6, February 18, 2021 [2021-02-18], pages 3060-3070, XP055873969, ISSN: 0740-3194, DOI: 10.1002 / mrm.28725).

[0022] Goora et al. propose directly measuring the temporal evolution of a magnetic field gradient and extracting a system impulse response. The system impulse response is used to determine optimized input waveforms that produce a desired gradient response at a system output (Goora Frederic G et al.: "Arbitrary magnetic field gradient waveform correction using an impulse response based pre-equalization technique," JOURNAL OF MAGNETIC RESONANCE, ACADEMIC PRESS, ORLANDO, FL, US, Vol. 238, November 15, 2013 [2013-11-15], pages 70-76, XP028670722, ISSN: 1090-7807, DOI: 10.1016 / J.JMR.2013.11.003).

[0023] It is an object of the present invention to provide a method for correcting deviations between played out gradient pulses from expected gradient curves during a magnetic resonance measurement.

[0024] This object is achieved according to the invention by the subject matter of the independent patent claims. Advantageous embodiments and expedient further developments are the subject matter of the dependent claims.

[0025] The method according to the invention for correcting an effect of a disturbance on a gradient system of a magnetic resonance device during a magnetic resonance measurement comprises the following steps: Emitting a gradient pulse g ref (t) by means of an amplifier of the gradient system, detecting a gradient profile g real (t) which is established by the emission of the gradient pulse g ref (t) in an examination region of the magnetic resonance device by means of a magnetic field measurement in the examination region, detecting an output signal i out (t) of the amplifier for the gradient pulse g ref (t), determining a gradient system transfer function (GSTF, GIRF) as a function of the detected gradient profile g real (t) and the detected output signal i out (t) of the amplifier, determining an output signal i cor (t) of the amplifier for which the gradient system provides an expected gradient profile g nom (t), as a function of the determined gradient system transfer function (GSTF, GIRF), and Adjusting an input signal i in (t) of the amplifier depending on the determined output signal i cor (t) of the amplifier and the detected output signal i out (t) of the amplifier.

[0026] In one step of the method according to the invention, at least one gradient pulse g ref (t) is emitted. The gradient pulse g ref (t) can represent part of a pulse sequence, which can comprise one or more gradient pulses, but also one or more RF pulses. The gradient pulse can, for example, represent a regular gradient pulse of the pulse sequence required to perform a magnetic resonance measurement of an examination subject. For example, a regular gradient pulse can comprise an imaging gradient or a diffusion gradient. An imaging gradient preferably serves to localize the effect of an RF pulse. However, the gradient pulse can also be emitted exclusively as a test gradient pulse, particularly in cases in which other gradient pulses of a pulse sequence are unsuitable as test gradient pulses.Furthermore, the gradient pulse can serve a dual function as a test gradient pulse and as a regular gradient pulse in a pulse sequence. A test gradient pulse can, in particular, be a gradient pulse that is used directly or indirectly to correct the effect of a disturbing effect on the gradient system of the magnetic resonance device during a magnetic resonance measurement. Disturbing effects, such as the occurrence of eddy currents, timing and gain errors, field fluctuations, variable temperature influences, and / or the non-linear behavior of hardware components, such as... z. B. the amplifier and / or the gradient coil of the gradient system, can cause a deviation between the output gradient pulses and the expected gradient curves.

[0027] A magnetic resonance measurement preferably represents an imaging examination of an examination subject, in particular a patient. It is conceivable that two-dimensional, three-dimensional, and / or time-dependent three-dimensional image data of the examination subject are acquired during the magnetic resonance measurement. However, it is equally conceivable that a magnetic resonance measurement is performed for a calibration and / or test of the magnetic resonance device, but also a magnetic resonance spectroscopy measurement. In this case, the examination subject can comprise, for example, a reference object, a phantom, and / or a field sample, which interact in a predetermined manner with a pulse sequence played out during the magnetic resonance measurement.In one embodiment, the implementation of the magnetic resonance measurement comprises the objective of characterizing deviations between the gradient pulses played out and the expected gradient profiles by means of the method according to the invention.

[0028] The emitted gradient pulse g ref (t) preferably has a triangular, rectangular, or trapezoidal shape. Gradient pulses designed in this way cover a broad frequency spectrum and are therefore particularly suitable for correcting trajectories throughout k-space. However, it is also conceivable for the emitted gradient pulse to have a radial or spiral trajectory.

[0029] Preferably, an expected or nominal gradient profile g nom (t) is associated with the emitted gradient pulse g ref (t). The expected gradient profile can coincide with the emitted gradient pulse if no interference effects occur in the gradient system. In this case, the emitted gradient pulse can be a "nominal gradient pulse," which is characterized by nominal field strength values ​​that are to be expected when a gradient pulse is emitted. However, in a real magnetic resonance measurement, the nominal field strength values ​​will deviate from the actual field strength values ​​of the gradient pulse due to dynamic interference. This deviation can be determined by a corresponding measurement of the gradient profile g real (t).

[0030] In one step, the gradient profile g real (t) is recorded by means of a magnetic field measurement in an examination region of the magnetic resonance device. The recorded gradient profile is, in particular, actual field strength values ​​which are generated in the examination region by the emitted gradient pulse g ref (t). The gradient profile can be recorded, for example, using a magnetometer or a field camera. The magnetometer or the field camera can be introduced into the examination region of the magnetic resonance device to record the gradient profile. For this purpose, the field camera can have a signal connection to an input interface of the magnetic resonance device and transmit the recorded gradient profile to the input interface via the signal connection. It is also conceivable that the gradient profile is recorded by a radio-frequency receiving antenna of the magnetic resonance device.For this purpose, a field sample or a volume of a reference substance, such as water or oil, can be introduced into the examination region of the magnetic resonance device and excited by a predetermined RF pulse. A magnetic resonance signal of the field sample or reference substance received by a radio-frequency receiving antenna of the magnetic resonance device can then be analyzed to obtain information about a property of the emitted gradient pulse. The analysis of the received magnetic resonance signal can, in particular, comprise a comparison of an expected magnetic resonance signal of the field sample or reference substance with the received magnetic resonance signal.

[0031] A gradient pulse can be understood as a magnetic field strength or a time-dependent magnetic field strength curve, which is to be provided by the gradient system. A gradient curve, on the other hand, can be characterized by a measured field strength or a recorded, time-dependent field strength curve. The gradient curve is thus to be understood in particular as an actual magnetic field gradient that is established by the emission of the gradient pulse in the examination area. Due to the effect of the above-mentioned interference effects, the gradient curve may deviate from the gradient pulse.

[0032] In a further step, an output signal i out (t) of the amplifier for the gradient pulse g ref (t) is detected. The output signal i out (t) of the amplifier can in particular be an electrical current which is fed from the amplifier into a gradient coil of the gradient system. The output signal of the amplifier is preferably detected while the gradient pulse g ref (t) is being emitted by a gradient coil of the gradient system. The output signal of the amplifier can be detected using a suitable measuring device, such as a current clamp, a Rogowski coil, a voltmeter, or even a measuring or current transformer. The detected output signal of the amplifier is preferably transmitted to the input interface of the magnetic resonance device.However, it is also conceivable that the amplifier's output signal is determined as a function of any desired measured variable and / or with the aid of a known characteristic curve or operating characteristic of the amplifier. The term amplifier is understood in particular to mean a gradient amplifier, which amplifies and / or transforms a signal with a low signal level (input signal of the amplifier) ​​to a signal with a high level (output signal of the amplifier). The amplifier's output signal can be fed to a gradient coil of the gradient system, which accordingly generates a magnetic field gradient in the examination region (emitted gradient pulse). In a preferred embodiment, the gradient system has at least three gradient coils, which are designed to provide magnetic field gradients aligned orthogonally to one another.

[0033] In one step of the method according to the invention, a gradient system transfer function (GSTF, GIRF) is determined as a function of the detected gradient profile g real (t) and the detected output signal i out (t) of the amplifier. The gradient system transfer function can be determined using a computing unit of the magnetic resonance device, which has a signal connection to the input interface. For simplicity, the two terms GSTF and GIRF will be subsumed below under the term gradient system transfer function, since the two functions can be converted into one another using equation (1).

[0034] The gradient system transfer function is usually determined as a function of the quotient of a gradient G real (f) generated by the amplifier and an expected or nominal gradient G nom (f) according to equation (4). This type of gradient system transfer function represents a so-called gradient input GSTF, in which the nominal input gradient G nom (f) is used to determine the gradient system transfer function. GSTF = G real f G nom f

[0035] In contrast, the gradient system transfer function in the method according to the invention is determined as a function of the detected gradient profile g real (t) and the detected output signal i out (t) of the amplifier according to equation (5). For this purpose, the detected output signal i out (t) of the amplifier and the detected gradient profile g real (t) can first be transformed into the frequency domain (I out (f), G real (f)) using Fourier transformation. GSTF = G real f I out f

[0036] Preferably, the gradient system transfer function GSTF is determined using a plurality of different gradient pulses to cover a broad frequency spectrum. The covered frequency spectrum can, for example, comprise a frequency range between -20 kHz and 20 kHz, in particular between -10 kHz and 10 kHz.

[0037] In a further step of the method according to the invention, an output signal i cor (t) of the amplifier, for which the gradient system provides an expected gradient profile (g nom (t)), is determined as a function of the determined gradient system transfer function (GSTF, GIRF). The output signal i cor (t) of the amplifier is preferably determined by means of the computing unit of the magnetic resonance apparatus. The computing unit preferably has a signal connection to the input interface. As described above, the input interface can be designed to receive the gradient profile and the output signal of the amplifier and to transmit them to the computing unit via the signal connection. The signal connection can be wireless or wired.

[0038] The method according to the invention advantageously compensates for interference effects, such as non-linear operating characteristics of the amplifier and / or variable temperature influences on the gradient system. Furthermore, image artifacts due to such interference effects can be reduced or avoided, and the quality of acquired image data can be advantageously increased. In particular, the output of the amplifier's input signal enables direct adjustment of the amplifier. This advantageously avoids computationally intensive and / or time-consuming correction of interference effects during image data reconstruction.

[0039] In a preferred embodiment of the method according to the invention, the transmission of the gradient pulse g ref (t) and the acquisition of the gradient profile g real (t) take place in a repetitive manner, in particular with a frequency between one and five repetition intervals of a pulse sequence of the magnetic resonance measurement.

[0040] The transmission of the gradient pulse and the acquisition of the gradient profile preferably occur repeatedly at discrete time intervals. It is conceivable that the repeated transmission of the gradient pulse and the acquisition of the gradient profile occur at a frequency between one and five repetition intervals of the pulse sequence of the magnetic resonance measurement.

[0041] In one embodiment, the frequency of repeated transmission of the gradient pulse and the acquisition of the gradient profile during the pulse sequence of the magnetic resonance measurement is reduced to a minimum. Experiments indicate that relevant temperature changes during the magnetic resonance measurement can occur after approximately five repetition intervals or can only be quantified by measurement after a period of five repetition intervals. By reducing the frequency of repeated transmission of the gradient pulse and the acquisition of the gradient profile, the time required for the magnetic resonance measurement and / or the computational effort for correcting interference effects can be advantageously reduced.

[0042] In a preferred embodiment, the gradient pulse is repeatedly transmitted and the gradient profile is recorded in each repetition interval. This advantageously allows for prompt compensation of even unexpectedly rapidly occurring interference effects that affect the magnetic field gradients. A repetition interval can be understood as the time period that elapses between two consecutive excitation pulses, in particular two consecutive RF pulses.

[0043] In one embodiment, the method according to the invention further comprises the step: Determining a temperature of a gradient coil of the gradient system, where determining the output signal i cor (t) of the amplifier depends on the temperature of the gradient coil.

[0044] The gradient coil from which the temperature is determined corresponds to the gradient coil which emits the gradient pulse g ref (t) as a function of the amplifier's output signal i out (t). The temperature of a gradient coil can be determined, for example, using one or more temperature sensors positioned on the gradient system, in particular on the gradient coil. The temperature sensor is preferably mechanically connected to the gradient coil and has a thermal coupling with the gradient coil via the mechanical connection. The temperature sensor can be designed to transmit a measured value for the temperature of the gradient coil or a signal indicative of the temperature of the gradient coil to the computing unit of the magnetic resonance device via a signal connection.The computing unit can be configured to determine the output signal of the amplifier as a function of the temperature of the gradient coil. It is also conceivable that the temperature of the gradient coil is determined as a function of a measured value and / or an operating characteristic of the amplifier that differ from the measured temperature value.

[0045] According to a further embodiment, the method according to the invention further comprises the step: Determining a resistance and / or inductance of the gradient coil, wherein the temperature of the gradient coil is determined as a function of the resistance and / or inductance of the gradient coil.

[0046] Preferably, the temperature of the gradient coil is determined as a function of a resistance and / or an inductance of the gradient coil or the gradient coils. Of course, other measured values ​​can also be used to determine the temperature of the gradient coil. Preferably, the temperature of the gradient coil is determined as a function of a measuring resistor (shunt resistor) connected to a voltmeter. However, it is also conceivable that the temperature of the gradient coil is determined as a function of a temperature-dependent property, such as z. B. a length of a reference wire, but also an inductance of the gradient coil.

[0047] In a further embodiment of the method according to the invention, the temperature of the gradient coil is determined as a function of a model and / or an intelligent algorithm.

[0048] An intelligent algorithm can include any application of artificial intelligence. An artificial intelligence can z. B. a self-learning algorithm, a neural network, a multi-layer neural network, an expert system, or the like, which determines the temperature of the gradient coil depending on one or more measured values ​​of the gradient system. It is also conceivable that the intelligent algorithm determines the temperature of the gradient coil depending on a known operating characteristic of the amplifier and / or the gradient coil. The intelligent algorithm can further use a model of the gradient coil and / or the amplifier to determine the temperature of the gradient coil.

[0049] Furthermore, the temperature of the gradient coil can be determined depending on a model. A model can be, in particular, a simulation model, an analytical model, an empirical model, and / or a model-based approach that describes or models the amplifier and / or the gradient coil.

[0050] In a preferred embodiment, the temperature of the gradient coil is determined as a function of the detected output signal i out (t) of the amplifier.

[0051] Knowing the current fed into the gradient coil by the amplifier allows the temperature to be determined very precisely based on fundamental physical relationships and / or analytical models of the gradient coil. The temperature of the gradient coil is preferably determined using a model and / or an intelligent algorithm within the computer unit of the magnetic resonance system. Of course, determining the temperature of the gradient coil can also involve determining the temperatures of multiple gradient coils used in a pulse sequence.

[0052] By determining the temperature of the gradient coil, the amplifier's output signal can advantageously be determined with increased accuracy. Furthermore, the amplifier's output signal can be determined as a function of the gradient coil's temperature in a particularly robust and / or reproducible manner, since variable temperature influences on the gradient coil can be taken into account.

[0053] Furthermore, the method according to the invention comprises the step: Adjusting an input signal i in (t) of the amplifier depending on the determined output signal i cor (t) of the amplifier and the detected output signal i out (t) of the amplifier.

[0054] It is conceivable that the amplifier's input signal is adjusted as a function of a suitable control device. Such a control device can, for example, be designed to adjust the amplifier's input signal as a function of the determined amplifier output signal and the detected amplifier output signal. The control device can, in particular, be designed to adjust the amplifier's input signal as a function of a difference between the determined amplifier output signal and the detected amplifier output signal.

[0055] Preferably, the amplifier's input signal is adjusted continuously or in discrete time steps. For example, the amplifier's input signal is adjusted multiple times per repetition interval, once per repetition interval, every second repetition interval, every third repetition interval, or every fourth or fifth repetition interval. The amplifier's input signal can be adjusted, in particular, after each transmission of a gradient pulse g ref (t) and / or detection of a gradient profile g real (t).

[0056] It is also conceivable that the amplifier input signal is adjusted in a second repetition interval depending on an amplifier output signal determined in a first repetition interval or depending on a gradient system transfer function determined in the first repetition interval. The first repetition interval and the second repetition interval can also be separated by several repetition intervals. This procedure is also referred to as "pre-emphasis." With this procedure, a correction can already be made during the transmission of the gradient pulse in the second repetition interval, so that image reconstruction no longer requires corrections to the gradient fields.

[0057] By adapting the input signal of the amplifier according to the invention, interference effects in the gradient system can be corrected in a timely and / or automatic manner in an advantageous manner.

[0058] According to a further embodiment of the method according to the invention, the input signal i in (t) of the amplifier is adapted by means of an adaptive filter, in particular a Volterra filter, a spline filter and / or a kernel filter.

[0059] Preferably, the input signal of the amplifier can be adaptively adjusted using a non-linear, adaptive filter, so that interference effects in the gradient system are compensated by changing the input signal of the amplifier. In this case, the input signal applied to an input of the amplifier can pass through the adaptive filter and form a so-called output sequence. This output sequence can be compared with a sequence to be formed by the adaptive filter, such as the determined output signal i cor (t) of the amplifier, in order to determine an error signal. If the output sequence deviates from the sequence to be formed, a control algorithm of the adaptive filter can change filter coefficients in order to minimize the error signal and adapt the output sequence of the adaptive filter to the sequence to be formed. Known methods for minimizing the squared errors, such as, for example,the Least Mean Squares algorithm (LMS) or the Recursive Least Squares algorithm (RLS).

[0060] Preferably, the adaptive filter comprises a Volterra-LMS method in which discrete input values ​​are replaced by various non-linear algebraic expressions, so-called Volterra series.

[0061] The amplifier's input signal can be adjusted, for example, directly or indirectly depending on an output sequence of the adaptive filter. In one embodiment, the output sequence of the adaptive filter is transmitted as a control signal via a signal connection to the amplifier and / or a control unit of the amplifier or the magnetic resonance device. The output sequence can, in particular, represent the amplifier's input signal. However, it is also conceivable for the output sequence to represent a correction signal, depending on which the amplifier's input signal is adjusted.

[0062] The adaptive filter can be implemented, for example, as an analog filter or a digital filter. In a preferred embodiment, the adaptive filter is implemented as part of a DSP (digital signal processor), an FPGA (field programmable gate array), and / or an FPAA (field programmable analog array).

[0063] By using an adaptive filter, even complex and / or non-linear interference effects of the gradient system can be advantageously compensated.

[0064] In a further embodiment of the method according to the invention, the transmission of the gradient pulse g test (t) comprises a transmission of a plurality of gradient pulses g test,i (t), wherein the plurality of gradient pulses (g test,i (t)) covers a frequency spectrum between -20 kHz to +20 kHz, in particular -10 kHz to +10 kHz.

[0065] Preferably, the detection of the gradient profile g real (t) also includes detecting a gradient profile for each gradient pulse of the plurality of gradient pulses. An output signal i out (t) of the amplifier can also be detected for each gradient profile. It is conceivable that a gradient system transfer function for a plurality of gradient pulses and a plurality of amplifier output currents is determined in this way. The gradient system transfer function can be characterized by a superposition of the plurality of detected gradient profiles and detected amplifier output currents.

[0066] By determining the gradient system transfer function for a plurality of gradient pulses covering a broad frequency spectrum, interference effects of the gradient system for any gradient pulses encompassed by the frequency spectrum can be advantageously corrected or compensated.

[0067] The correction device according to the invention has an output interface configured to emit at least one gradient pulse g ref (t). The output interface may, for example, comprise an output channel of an amplifier of a gradient system of the magnetic resonance system.

[0068] The correction device according to the invention further comprises an input interface which is designed to receive a gradient profile g real (t) acquired by means of a magnetic field measurement in an examination region of a magnetic resonance device, which gradient profile is established by emitting the gradient pulse g ref (t) in the examination region, as well as an output signal i out (t) of an amplifier for the gradient pulse g ref (t). For this purpose, the input interface preferably has a signal connection to a field camera, a magnetometer or a radio-frequency receiving antenna. The input interface can also be designed as part of the radio-frequency receiving antenna. It is also conceivable for the input interface to have a signal connection to a current sensor, a voltage sensor and / or a temperature sensor.In principle, the input interface can have a signal connection to any sensor configured to metrologically detect a property of the gradient system according to an embodiment described above. Furthermore, the correction device according to the invention has a computing unit configured to determine a gradient system transfer function (GSTF, GIRF) as a function of the detected gradient profile g real (t) and the detected output signal i out (t) of the amplifier. The computing unit is further configured to determine an output signal i cor (t) of the amplifier as a function of the gradient system transfer function.

[0069] The correction device according to the invention is further configured to adapt an input signal i in (t) of the amplifier as a function of the determined output signal i cor (t) of the amplifier and the detected output signal i out (t) of the amplifier. The determination of the gradient system transfer function GSTF, the output signal i cor (t) of the amplifier, and also the adaptation of the input signal i in (t) of the amplifier can be carried out by means of the correction device according to an embodiment of the method according to the invention described above.

[0070] The correction device according to the invention shares the advantages of the method according to the invention for correcting an effect of a disturbance effect on a gradient system of a magnetic resonance device during a magnetic resonance measurement.

[0071] The magnetic resonance apparatus according to the invention comprises a correction device according to an embodiment described above, wherein the magnetic resonance apparatus is designed to adapt an input signal i in (t) of an amplifier during a magnetic resonance measurement of an examination object by means of the correction device in order to provide an expected gradient profile g nom (t).

[0072] In particular, the magnetic resonance apparatus according to the invention can be configured to autonomously coordinate and execute a sequence of individual method steps of a method according to the invention according to an embodiment described above. In this way, the adjustment of the amplifier's input signal as a function of a gradient system transfer function and a determined amplifier output signal can advantageously be performed reproducibly and / or automatically.

[0073] The magnetic resonance system according to the invention comprises a basic field magnet system, a radio-frequency transmitting antenna, a gradient system, a radio-frequency receiving antenna and a correction device according to an embodiment described above, as well as a control unit for controlling the basic field magnet system, the radio-frequency transmitting antenna, the gradient system and the radio-frequency receiving antenna.

[0074] By providing the magnetic resonance system according to the invention, the components of the magnetic resonance system according to the invention, such as the basic field magnet system, the radio-frequency transmitting antenna, the gradient system, the radio-frequency receiving antenna as well as the correction device and the control unit, can be advantageously coordinated with one another, so that a time-efficient and / or robust implementation of a method according to the invention according to an embodiment described above is enabled.

[0075] The computer program product according to the invention comprising a computer program which can be loaded directly into a memory unit of the control unit of a magnetic resonance system according to the invention, comprising program sections for executing all steps of a method according to the invention according to an embodiment described above when the computer program is executed by means of the control unit of the magnetic resonance system.

[0076] The correction device according to the invention can preferably be implemented in the form of software on a suitable programmable control unit of a magnetic resonance system with a corresponding memory unit. A radio-frequency transmission unit, an output interface, and a radio-frequency reception unit can also be implemented at least partially in the form of software units, although other units of these components can in turn be purely hardware units. Such hardware units can be, for example, a radio-frequency amplifier, an antenna device of the radio-frequency transmission unit and / or the radio-frequency reception unit, a gradient pulse generation device of the output interface, an analog / digital converter of the radio-frequency reception unit, and the like.

[0077] A largely software-based implementation, in particular of the components mentioned, has the advantage that even previously used control units of a magnetic resonance system can be easily retrofitted by means of a software update in order to operate in the manner according to the invention. In this respect, the object is also achieved by a computer program product which is stored in a portable storage unit and / or made available for transmission via a network and can thus be loaded directly into a memory of a programmable control unit of the magnetic resonance system, with program sections in order to carry out all steps of the method according to the invention according to an embodiment described above when the program is executed by means of the control unit. Such a computer program product can, in addition to the computer program, optionally contain additional components such as, for example,include documentation and / or additional components, including hardware components such as hardware keys (dongles, etc.) for using the software.

[0078] The computer-readable medium according to the invention stores program sections that can be read and executed by a computing unit in order to carry out all steps of a method according to the invention according to an embodiment described above when the program sections are executed by the computing unit.

[0079] A computer-readable medium, such as a memory stick, a hard disk, or another portable or permanently installed data storage device, can be used to transport parts of the control unit and / or for storage on or in the control unit. The computer program sections that can be read and executed by a processing unit of the control unit are stored on the medium. For this purpose, the processing unit can, for example, have one or more cooperating microprocessors or the like.

[0080] Further advantages and details of the present invention will become apparent from the exemplary embodiments described below and from the drawings. Fig. 1 shows an embodiment of a magnetic resonance system according to the invention, Fig. 2 shows a schematic representation of the effect of interference effects during a magnetic resonance measurement, Fig. 3 shows a diagram illustrating a variable temperature influence on a gradient system of a magnetic resonance system, Fig. 4 shows a schematic representation of a principle of adapting an input signal of an amplifier by means of a correction device according to the invention, Fig. 5 shows a diagram illustrating a determination of a gradient system transfer function as a function of a plurality of gradient pulses, Fig. 6 shows a flowchart of a possible embodiment of a method according to the invention.

[0081] In the Fig. 1 The magnetic resonance system 1 shown is a magnetic resonance device 2 with an examination region 3 designed as a patient tunnel, into which an examination subject can be completely inserted. In the present example, the examination subject is a patient who can be positioned relative to the magnetic resonance device 2 by means of the patient support device 8. In principle, however, the magnetic resonance system 1 according to the invention can also comprise a different magnetic resonance device, e.g., with a C-shaped housing that is open at the side. It is essential that the magnetic resonance device 2 is designed to perform magnetic resonance measurements in order to be able to acquire image data of the examination subject.

[0082] The magnetic resonance apparatus 2 is preferably equipped with a basic field magnet system 4, at least one gradient coil 6, a radio-frequency transmitting antenna 5, and a radio-frequency receiving antenna 7. In the illustrated embodiment, the radio-frequency transmitting antenna 5 is a body coil permanently installed in the magnetic resonance apparatus 2. In contrast, the radio-frequency receiving antenna 7 is one or more local coils arranged on the patient. In principle, however, the body coil can also be designed as a radio-frequency receiving antenna and the local coils as a radio-frequency transmitting antenna and / or have a corresponding functionality. The basic field magnet system 4 is designed in the present case to generate a basic magnetic field in the longitudinal direction of the patient, i.e., along the longitudinal axis of the magnetic resonance apparatus 2 running in the Z direction. The gradient system 9 preferably comprises (see Fig. 4 ) of the magnetic resonance device 2 a plurality of individually controllable gradient coils 6 in order to be able to play out gradients in the X, Y or Z direction independently of one another.

[0083] The magnetic resonance system 1 has a central control unit 13, which is used to control the magnetic resonance system 1. This central control unit 13 can also include a sequence control unit 14. The sequence control unit 14 controls the sequence of RF pulses and gradient pulses as a function of a selected pulse sequence PS or a sequence of multiple pulse sequences PS for acquiring image data of a relevant volume of the examination object, in particular one or more slices of the volume region of interest, within the framework of a magnetic resonance measurement. Such a pulse sequence PS can, for example, be specified and parameterized within a measurement or control protocol P.Typically, various control protocols P for different magnetic resonance measurements are stored in a storage unit 19 and can be selected by an operator, modified as needed, and then used to perform the magnetic resonance measurement. In the present case, the control unit 13 contains several pulse sequences PS for acquiring raw data.

[0084] To output individual RF pulses of a pulse sequence PS, the central control unit 13 has a transmitting unit 15. The transmitting unit 15 generates and amplifies the RF pulses and transmits them via a signal connection and a suitable interface to the radio-frequency transmitting antenna 5 (not shown in detail). To control the gradient coils 6 of the gradient system 9, in particular to switch or play out the gradient pulses appropriately according to a predetermined pulse sequence PS, the control unit 13 has an output interface 16. By means of the output interface 16, for example, diffusion gradient pulses or spoiler gradient pulses can be applied. In the present case, the output interface 16 has a signal connection to a gradient coil 6 (see Fig. 4 ) of the gradient system 9, which generates corresponding magnetic field gradients in the examination region 3 as a function of the applied gradient pulses. The sequence control unit 14 communicates in a suitable manner, e.g., by transmitting sequence control data SD, with the transmitting unit 15 and the output interface 16 for executing the pulse sequence PS.

[0085] The control unit 13 also has a radio-frequency receiving unit 17 (which also communicates in a suitable manner with the sequence control unit 14) in order to receive magnetic resonance signals in a coordinated manner by means of the radio-frequency receiving antenna 7 within a readout window predetermined by the pulse sequence PS and thus to acquire the raw data.

[0086] The magnetic resonance system 1 can further comprise a reconstruction unit 120a, which is configured to reconstruct image data of the examination subject as a function of acquired magnetic resonance signals. This reconstruction can be performed based on parameters that can be specified in the respective measurement or control protocol P. The reconstructed image data can then be stored, for example, in the storage unit 19.

[0087] The correction device 110 has a signal connection with the other components, in particular the output interface 16 and / or the sequence control unit 14. Alternatively, the correction device 110 can also be configured as a part of the sequence control unit 14. The correction device 110 is preferably designed to generate suitable gradient pulses g ref (t) and to transmit them to the output interface 16 or the gradient coil 6. It is conceivable that the correction device 110 has an amplifier 113 (see Fig. 4 ) of the gradient system 9 and specifies an input signal i in (t) of the amplifier 113. Alternatively, the correction device 110 can also be separate from the amplifier 113 and initiate a transmission of gradient pulses g ref (t) by means of a suitable signal connection. The correction device 110 is in particular designed to adapt an input signal i in (t) of the amplifier 113 as a function of a determined output signal i cor (t) and a detected output signal i out (t) of the amplifier 113 (see Fig. 4 ).

[0088] The central control unit 13 can be operated via a terminal 11 with an input unit 10 and a display unit 12. The magnetic resonance system 1 can preferably be controlled by an operator using the terminal 11. The display unit 12 can be configured to output acquired image data from a magnetic resonance measurement. The input unit 10 can be configured, optionally in combination with the display unit 12, to parameterize and start magnetic resonance measurements using the magnetic resonance system 1. This can, in particular, comprise a selection and / or modification of control protocols P and / or pulse sequences PS.

[0089] The magnetic resonance system 1 according to the invention and in particular the control unit 13 can also have a plurality of further components, not shown in detail here but usually present in such systems, such as a network interface in order to connect the entire system to a network and to be able to exchange and / or process raw data and / or image data or parameter maps, but also other data, such as patient-relevant data or control protocols.

[0090] The detailed procedure for acquiring suitable magnetic resonance signals by applying RF pulses and transmitting gradient pulses, and for reconstructing image data or parameter maps from them, is generally known to those skilled in the art. Likewise, various measurement sequences, such as EPI measurement sequences or other measurement sequences for generating image data, including diffusion-weighted image data, are generally known to those skilled in the art. Therefore, these aspects will not be discussed in detail below.

[0091] In Fig. 2 A comparative representation 30 of emitted gradients G ref and the actually played back gradients G real of a pulse sequence of a magnetic resonance measurement is shown. The emitted gradients G ref have field strength values ​​that are determined by the control unit 13 of the magnetic resonance system 1. In contrast, the real gradients G real correspond to the field strength values ​​actually measured in an examination region. In a partial representation 30a, a diagram of an input gradient or emitted gradient G ref is shown, which is generated by the control unit 13 of the magnetic resonance device 2 and transmitted to the gradient coil 6 of the magnetic resonance device 2 of the magnetic resonance system 1. The gradient G real actually played back by the gradient coil 6 is shown in a partial diagram 30c. The emitted gradient G ref has in the Fig. 2 shown example has a triangular shape. Such a triangular shape covers a relatively wide frequency range and can therefore represent the response behavior of a gradient in the frequency domain relatively well. The gradient G real actually played out has a certain degree of distortion due to the interference effects of the gradient system 9 described above, such as a non-linear behavior of an amplifier 113 (see Fig. 4 ), thermal changes of the hardware components, but also eddy currents and / or timing and amplifier errors, a gradient shape deviating from the emitted gradient G ref .

[0092] In a partial diagram 30d, both gradients G ref and G real are shown together. As can be seen in partial diagram 30e, the real gradient G real can also exhibit a time offset relative to the transmitted gradient G ref.

[0093] A section of sub-graph 30d is shown enlarged in sub-graph 30e. In sub-graph 30e, it can be seen that the actual gradient G real, or an amplitude A of the actual gradient G real, falls below the emitted gradient G nom at t = 122 s and assumes a higher magnitude value than the emitted gradient G ref at approximately t = 130 s, only to coincide again with the emitted gradient at the zero line at values ​​of t = 160 s.

[0094] In Fig. 3 A diagram 30 is shown, which represents the magnitudes M of two gradient system transfer functions GSTF at different reference temperatures. The gradient system transfer function GSTF is determined here using the quotient of the real gradient G real and the detected output signal I out of the amplifier 113. In Fig. 3 Two gradient system transfer functions GSTF 20 , GSTF 40 are illustrated for two different mean temperatures of the gradient system 9 of 20°C and 40°C. The magnitudes M of the GSTF 20 , GSTF 40 are plotted against the frequency f in kHz. As can be seen in Figure 30, the magnitude M of the gradient system transfer function is slightly higher at 40°C than at 20°C.

[0095] Fig. 4 shows a schematic representation of a process for adapting the input signal i in (t) of the amplifier 113 as a function of the determined output signal i cor (t) and the detected output signal i out (t) of the amplifier 113. The gradient system 9 of the magnetic resonance system 1 preferably has a sensor 115 which is designed to detect a signal which is correlated with the output signal i out (t) of the amplifier 113. The sensor 115 can further be designed to transmit the signal to a computing unit 18 of the magnetic resonance system 1 by means of a signal connection.

[0096] The magnetic resonance system 1 further comprises a field camera 116 which is designed to detect the gradient profile g real (t) of the emitted gradient pulse g ref (t) and to transmit it to the computing unit 18 by means of a signal connection.

[0097] The computing unit 18 is accordingly designed to determine the gradient system transfer function GSTF as a function of the real gradient profile g real (t) and the detected output signal i out (t) of the amplifier 113. The computing unit 18 is also designed to determine the output signal i cor (t) of the amplifier 113, at which the gradient system 9 provides the expected gradient profile g ref (t), as a function of the gradient system transfer function GSTF. The output signal i cor (t) of the amplifier 113 thus represents, in particular, a target value of an output signal of the amplifier 113, at which interference effects of the gradient system 9 are corrected or compensated.

[0098] In the present example, the computing unit 18 transmits the output signal i cor (t) of the amplifier 113 and the detected output signal i out (t) of the amplifier 113 to the correction device 110. It is also conceivable for the computing unit 18 to be integrated into the correction device 110. The correction device 110 is designed to adapt the input signal i in (t) of the amplifier 113 as a function of the output signal i cor (t) of the amplifier 113 and the detected output signal i out (t) of the amplifier 113. For this purpose, the correction device 110 can, for example, have a control device, in particular an adaptive filter, according to an embodiment described above.

[0099] In the example shown, the recorded gradient G real (f) agrees with the emitted or nominal gradient G ref (f) (as in Fig. 2 shown), since the correction device 110 corrects the interference effects in the gradient system 9 by adjusting the input signal i in (t) of the amplifier 113. The course of the output gradient G ref (f) thus essentially corresponds to the expected or nominal gradient G nom (f).

[0100] In one embodiment, the temperature of the gradient coil 6 is detected by a temperature sensor 117 positioned on the gradient coil 6. It is conceivable that the temperature of the gradient coil 6 is transmitted to the input interface 112 and used by the computing unit 18 when determining the output signal i cor (t) of the amplifier 113.

[0101] In Fig. 5 A diagram 31 is shown which illustrates the principle of determining a gradient system transfer function GSTF using a combination of twelve different triangular gradient pulses g ref,i (t). The gradient pulses g ref,i (t) each have an output signal i out,i (t) output by the amplifier 113, which is transmitted to the gradient coil 6 via the output interface 16. Partial diagram 31a shows an ensemble of twelve triangular output signals i out,i (t) of the amplifier 113 with different line patterns (dashed, solid, dotted). When passing the gradient coil 6 of the gradient system 9 of the magnetic resonance apparatus 2, these generate actual gradient curves g real,i (t), which are shown in partial diagram 31b.As can be seen in the partial diagram 31b, the real gradient pulses g real,i (t) deviate from expected gradient curves g nom,i (t) due to interference effects, which essentially correspond to the emitted gradient pulses g ref,i (t) (see also . Fig. 2 ). The two sub-diagrams 31a and 31b are shown in the time domain. In contrast, sub-diagrams 31c and 31d show the output signals I out,i (f) of amplifier 113 and the real gradients G real,i (f) in the frequency domain. The output signals I out,i (f) of amplifier 113 and the real gradients G real,i (f) are each obtained by a Fourier transformation FT of i out,i (t) and g real,i (t) from the time domain to the frequency domain. Figure 31 also shows a gradient system transfer function GSTF, which is determined by dividing the real gradients G real,i (f) by the input signals I out,i (f) of amplifier 113. This fact is shown in Fig. 5 symbolized by a division sign "÷".

[0102] Fig. 6 shows a flowchart of an embodiment of a method according to the invention for correcting an effect of an interference effect on a gradient system 9 of a magnetic resonance device 2 during a magnetic resonance measurement.

[0103] In a step S1 of the method according to the invention, a gradient pulse g ref (t) is emitted by means of the amplifier 113. The gradient pulse can, in particular, represent a test gradient pulse or a regular gradient pulse during a magnetic resonance measurement of a patient. The gradient pulse is provided by the output interface 16 of the amplifier 113 of the gradient system 9 and transmitted to a gradient coil 6 of the gradient system 9, which provides the gradient pulse in the form of a magnetic field gradient in the examination region 3.

[0104] In one embodiment of the method according to the invention, the transmission of the gradient pulse g ref (t) and the acquisition of the gradient profile g real (t) take place in a repetitive manner, in particular with a frequency between one and five repetition intervals of the magnetic resonance measurement.

[0105] In a step S2, a gradient profile g real (t) is acquired by means of a magnetic field measurement in an examination region 3 of the magnetic resonance device 2. The gradient profile can be acquired, in particular, by means of a magnetometer or a field camera 116, which transmits information and / or acquired magnetic field data regarding the gradient profile to the input interface 112 of the magnetic resonance device 2, in particular to the correction device 110. For this purpose, the magnetometer or the field camera 116 can be connected to the input interface 112 via a wired or wireless signal connection.

[0106] In a further step S3, an output signal i out (t) of the amplifier 113 for the gradient pulse g ref (t) is detected. The output signal of the amplifier 113 is preferably detected by means of a suitable sensor 115, such as a current clamp, a Rogowski coil, a voltmeter, or even a measuring or current transformer. The sensor 115 can, in particular, have a signal connection to the input interface 112 in order to transmit information and / or a measured value relating to the output signal of the amplifier 113 to the control unit 13 and / or the computing unit 18.

[0107] In a step S4, a gradient system transfer function (GSTF, GIRF) is determined as a function of the detected gradient profile g real (t) and the measured output signal i out (t) of the amplifier 113. Preferably, the gradient system transfer function is determined according to equation (5) and the procedure described above (see Fig. 4 ).

[0108] In a further step S5, an output signal i cor (t) of amplifier 113 is determined, for which the gradient system provides an expected gradient profile g nom (t), depending on the determined gradient system transfer function (GSTF, GIRF). For this purpose, the output signal i cor (t) of amplifier 113 is preferably determined using equation (6). i cor t = F − 1 G nom f ⋅ GSTF − 1 f

[0109] The output signal i cor (t) of the amplifier 113 represents an output signal required to provide the expected, nominal gradient pulse g nom (t) in the examination region 3 of the magnetic resonance device 2. The nominal gradient pulse g nom (t) can essentially correspond to the gradient profile g real (t) as long as no interference effects occur in the gradient system 9 or any interference effects that occur are corrected by means of the correction device 110 by adjusting the input signal i in (t) of the amplifier 113. The term GSTF -1< represents the inverse of the GSTF, which is determined as a function of the acquired gradient profile g real (t) and the acquired output signal i out (t) of the amplifier 113 according to equation (5).

[0110] According to one embodiment, the output signal i cor (t) of the amplifier 113 is determined as a function of a temperature of a gradient coil 6. For this purpose, the temperature of the gradient coil 6 can be integrated into equation (7) for the nominal output signal i cor (t) of the amplifier 113. Preferably, an expression for the temperature-dependent gradient system transfer function GSTF T is used here. i cor T t = F − 1 G nom f ⋅ GSTF T − 1 f

[0111] The temperature dependence of the gradient system transfer function GSTF T can be described, for example, by means of a linear model for a temperature-dependent change of the gradient system transfer function Δ GSTF can be determined according to equation (8). Here, T meas can represent a matrix of measured temperature values ​​at different times T t=0 to T t=N, and m a model parameter. This approach can include heuristic modeling in which the model parameters m are calibrated. Δ GSTF = T meas ⋅ m

[0112] However, it is also conceivable that the temperature dependence of the gradient system transfer function GSTF T depends on a change in the resistance Δ R of the gradient coil 6. By changing the resistance of the gradient coils 6 at different temperatures, the temperature-dependent change of the gradient system transfer function Δ GSTF can be determined using the following relationship (9): Δ GSTF = f Δ R T

[0113] In an optional step S5.1, a temperature of a gradient coil 6 of the gradient system 9 is determined. Preferably, the temperature of one or more gradient coils 6 of the gradient system 9 is detected using one or more temperature sensors and transmitted via a signal connection to the control unit 13 and / or the computing unit 18 of the magnetic resonance system 1. The computing unit 18 is accordingly configured to determine the output signal i cor (t) of the amplifier 113 as a function of the temperature of the gradient coil 6 using equation (7).

[0114] In one embodiment, the temperature of the gradient coil 6 is determined based on a model and / or an intelligent algorithm. It is conceivable that the temperature of the gradient coil 6 is determined based on a known operating characteristic of the amplifier 113. Furthermore, an analytical model, an empirical model, and / or a simulation model of the amplifier 113, but also of the gradient coil 6, can be used to determine the temperature of the gradient coil 6. Furthermore, information about a mechanical and / or electrical structure of the amplifier 113 and / or the gradient coil 6, an arrangement and / or electrical connection of electronic components, and / or a thermal conductivity and a heat capacity of materials used can be used to determine the temperature of the gradient coil 6.

[0115] In an alternative embodiment, the temperature of the gradient coil 6 is determined as a function of the detected output signal i out (t) of the amplifier 113. Knowing the current fed into the gradient coil 6 by the amplifier 113, the temperature of the gradient coil 6 can be determined very precisely based on a simple analytical model of the gradient coil 6. Preferably, the temperature of the gradient coil 6 is determined using a model and / or an intelligent algorithm by means of the computing unit 18 of the magnetic resonance system 1.

[0116] In an optional step S5.2, a resistance and / or inductance of the gradient coil 6 of the gradient system 9 is determined. The determination of the resistance and / or inductance can be carried out in particular according to an embodiment of the method according to the invention described above. In one example, the temperature of the gradient coil 6 is determined as a function of a measuring resistor (shunt resistor) connected to a voltmeter. The resistance and / or inductance of the gradient coil 6 can then be used to determine the temperature of the gradient coil 6 in step S5.1.

[0117] In an optional step S6, an input signal i in (t) of the amplifier 113 is adjusted as a function of the determined output signal i cor (t) of the amplifier 113 and the detected output signal i out (t) of the amplifier 113. The adjustment of the input signal of the amplifier can be carried out directly or indirectly by means of a suitable control device, in particular by means of an adaptive filter. For adjusting the input signal of the amplifier, a correction device 110 is preferably used (see Fig. 4). This can receive an output signal i cor (t) of the amplifier 113, determined according to equation (6) or equation (7), from the computing unit 18 of the magnetic resonance system 1 and compare it in a quasi-continuous manner or in discrete time steps with the output signal i out (t) of the amplifier 113 detected by means of the input interface 112. If there is a deviation between the determined output signal i cor (t) and the detected output signal i out (t) of the amplifier 113, the input signal i in (t) of the amplifier 113 is adjusted by means of the correction device 110 such that the deviation is reduced. Thus, interference effects in the gradient system 6 can already be corrected by means of the correction device 110 when a gradient pulse is emitted.

[0118] In one embodiment, the input signal i in (t) of amplifier 113 is adapted using an adaptive filter, in particular a Volterra filter, a spline filter, and / or a kernel filter. The adaptive filter preferably comprises a Volterra-LMS method, in which a least-mean-square algorithm (LMS) is used to minimize squared errors.

[0119] It should be noted that the methods and structures described in detail above are merely exemplary embodiments and that the basic principle can be varied widely by those skilled in the art without departing from the scope of the invention, as defined by the claims. Furthermore, the described method is not limited to medical applications. For the sake of completeness, it should also be noted that the use of the indefinite articles "a" or "an" does not exclude the possibility that the relevant features may be present in multiple instances. Likewise, the term "unit" does not exclude the possibility that it consists of multiple components, which may also be spatially distributed.

Claims

1. Method for correcting an influence of an interference effect on a gradient system (9) of a magnetic resonance apparatus (2) during a magnetic resonance scan, having the steps: • emitting (S1) a gradient pulse (gref(t)) by means of an amplifier (113) of the gradient system (9), • capturing (S2) a gradient sequence (greal(t)) which becomes established in an examination region (3) of the magnetic resonance apparatus (2) by way of the emission of the gradient pulse (gref(t)), by means of a magnetic field measurement in the examination region (3), • capturing (S3) an output signal (iout(t)) of the amplifier (113) for the gradient pulse (gref(t)), • establishing (S4) a gradient system transfer function (GSTF, GIRF) dependent upon the captured gradient sequence (greal (t)) and the captured output signal (iout(t)) of the amplifier (113), characterised by the steps • establishing (S5) an output signal (icor(t)) of the amplifier (113) with which the gradient system (9) provides an expected gradient sequence (gnom(t)), dependent upon the established gradient system transfer function (GSTF, GIRF) and • adjusting (S6) an input signal (iin(t)) of the amplifier (113) dependent upon the established output signal (icor(t)) of the amplifier (113) and of the captured output signal (iout(t)) of the amplifier (113).

2. Method according to claim 1, wherein the emitting (S1) of the gradient pulse (gref(t)) and the capturing (S2) of the gradient sequence (greal(t)) take place in a repeating manner, in particular, with a frequency of between one and five repetition intervals of a pulse sequence (PS) of the magnetic resonance scan.

3. Method according to claim 1 or 2, further having the step: • establishing (S5.1) a temperature of a gradient coil (6) of the gradient system (9), wherein the establishing (S5) of the output signal (icor(t)) of the amplifier (113) takes place dependent upon the temperature of the gradient coil (6).

4. Method according to claim 3, further having the step: • establishing (S5.2) a resistance and / or an inductance of the gradient coil (6) of the gradient system (9), wherein the establishing (S5.1) of the temperature of the gradient coil (6) takes place dependent upon the resistance and / or the inductance of the gradient coil (6).

5. Method according to one of claims 3 or 4, wherein the establishing (S5.1) of the temperature of the gradient coil (6) takes place dependent upon a model and / or an intelligent algorithm.

6. Method according to one of claims 3 to 5, wherein the establishing (S5.1) of the temperature of the gradient coil (6) takes place dependent upon the captured output signal (iout (t)) of the amplifier (113).

7. Method according to one of the preceding claims, wherein the adjustment (S6) of the input signal (iin(t)) of the amplifier (113) takes place by means of an adaptive filter, in particular, a Volterra filter, a spline filter and / or a kernel filter.

8. Method according to one of the preceding claims, wherein the emission (S1) of the gradient pulse (gref(t)) comprises an emission of a plurality of gradient pulses (gref,i(t)), and wherein the plurality of gradient pulses (gtest,i(t)) covers a frequency spectrum of between -20 kHz and +20 kHz, in particular between -10 kHz and +10 kHz.

9. Correcting facility (110), having: • an output interface (16) which is designed to emit at least one gradient pulse (gref(t)), • an input interface (112) which is designed to capture a gradient sequence (greal(t)) captured by means of a magnetic field measurement in an examination region (3) of a magnetic resonance apparatus (2), which becomes established in the examination region (3) by way of the emission of the gradient pulse (gref(t)) and an output signal (iout(t)) of an amplifier (113) for the gradient pulse (gref(t)), characterised by • a computer unit (18) which is designed to establish a gradient system transfer function (GSTF, GIRF) and an output signal (icor(t)) of the amplifier (113) dependent upon the captured gradient sequence (greal(t)) and the captured output signal (iout(t)) of the amplifier (113), wherein the correcting facility (110) is designed to adjust an input signal (iin(t)) of the amplifier (113) dependent upon the established output signal (icor(t)) of the amplifier (113) and the captured output signal (iout(t)) of the amplifier (113).

10. Magnetic resonance system (1), comprising a main field magnet system (4), a radio frequency transmitting antenna (5), a gradient system (9), a radio frequency receiving antenna (7) and a correcting facility (110) according to claim 9 and a control unit (13) for actuating the main field magnet systems (4), the radio frequency transmitting antenna (5), the gradient system (9) and the radio frequency receiving antenna (7), wherein the magnetic resonance system (1) is designed to adjust an input signal (iin(t)) of an amplifier (113) during a magnetic resonance scan of an examination object by means of the correcting facility (110), in order to provide an expected gradient sequence (gnon(t)).

11. Computer program product having a computer program which can be loaded directly into a storage device (19) of the control unit (13) of a magnetic resonance system (1) according to claim 10, having program portions in order to carry out all the steps of a method according to one of claims 1 to 8 when the computer program is executed by means of the control unit (13) of the magnetic resonance system (1).

12. Computer-readable medium on which program portions that are able to be read in and executed by a computer unit (18) are stored, in order to carry out all the steps of a method according to one of claims 1 to 8, when the program portions are executed by the computer unit (18).