METHOD FOR OPERATING A MAGNETIC FIELD CAMERA

Description

[0001] The invention relates to a method for measuring a magnetic field distribution in a spatial volume to be measured using a magnetic resonance tomograph and a field camera for detecting a magnetic field distribution by means of a magnetic resonance measurement with a plurality of samples distributed over a spatial volume according to claim 1.

[0002] Magnetic resonance imaging scanners are imaging devices that, to create an image of a subject, align the nuclear spins of the subject with a strong external magnetic field and then excite them to precess around this alignment using an alternating magnetic field. The precession, or return, of the spins from this excited state to a lower-energy state, in turn generates a response alternating magnetic field, which is received via antennas.

[0003] Using magnetic gradient fields, a spatial coding is imprinted on the signals, which subsequently allows the received signal to be assigned to a volume element. The received signal is then evaluated, and a three-dimensional image of the object under examination is provided.

[0004] The quality of the generated images depends heavily on both the homogeneity of the static magnetic field and the linearity of the gradient fields used for spatial encoding. These are also influenced by dynamic effects such as eddy currents caused by rapidly changing gradient fields. Devices for spatially recording the magnetic fields are also called field cameras.

[0005] From the patent specification DE 10 2014 213 413, for example, a method and a device for measuring a magnetic field in a magnetic resonance imaging device with at least one field probe is known, to which at least one generated cancellation signal for reducing a residual magnetization in the at least one field probe (FS) can be applied.

[0006] From the document YING-HUA CHU ET AL: "Decoupled dynamic magnetic field measurements improves diffusion-weighted magnetic resonance images" (SCIENTIFIC REPORTS, Vol. 7, No. 1, September 14, 2017 (2017-09-14), XP055615098, DOI: 10.1038 / s41598-017-11138-8), field probes with localized NMR-active probes are known. They are suitable for monitoring magnetic fields. To reduce measurement errors caused by probe coupling, it is proposed to determine an asymmetric matrix with coupling coefficients of the field probes. The local signal of a probe is determined using the inverse matrix. This method is used to determine maps of dynamic magnetic fields in diffusion-weighted magnetic resonance imaging. The fields estimated using decoupled samples result in magnetic resonance images that are more robust against eddy currents that are oriented in different directions due to diffusion sensitivity gradients.

[0007] A method for decoupling field samples and improving the accuracy of dynamic magnetic field estimation is described in the document YING-HUA CHU ET AL.: "Accurate dynamic magnetic field monitoring and diffusion-weighted image reconstruction using uncorrelated local field measurements" (PROCEEDINGS OF THE INTERNATIONAL SOCIETY FOR MAGNETIC RESONANCE IN MEDICINE, ISMRM, 25TH ANNUAL MEETING AND EXHIBITION, HONOLULU, HI, USA, 22 APRIL - 27 APRIL 2017, No. 3915, 7 April 2017 (2017-04-07), XP040691483). A sensitivity matrix of field samples was determined to decouple magnetic resonance signals.

[0008] The existing field cameras have individual antenna coils on the field probes and separate receivers for evaluating the signals from the field probes, which, when a large number of field probes are used, represent a considerable effort to increase the spatial resolution.

[0009] The task therefore arises to make magnetic field measurement in a magnetic resonance imaging scanner simpler and more cost-effective.

[0010] The object is achieved by the inventive method according to claim 1.

[0011] The field camera used in the method according to the invention has a number of M samples. Samples are defined as volumes containing a medium whose nuclear spins exhibit magnetic resonance. In the simplest case, these can be, for example, cuvettes filled with water. The material of the cuvettes preferably has no or only slight diamagnetic or paramagnetic properties, such as Styrofoam, and is furthermore shaped as a spherical shell or ellipsoid to minimize field distortions of the magnetic field. The samples preferably have a small volume compared to the volume to be measured, for example, a volume smaller by a factor of 10, 50, 100, or 1000 than the volume to be measured.

[0012] The samples are distributed across the spatial volume to be measured. This means, for example, that the spatial volume is divided into M equally sized disjoint sub-volumes, with only one sample being arranged in each of the sub-volumes. It is also conceivable for the distance between two adjacent samples to be no smaller than a predetermined minimum distance. The minimum distance is preferably greater than the greatest extent of the sample in one spatial direction by a factor of more than 5, 10, 20, or 100. It can also be considered distributed if only one sample is arranged in each of the receiving volumes described below. It is also conceivable for the samples to be arranged on one or more concentric shells, for example spherical shells, around or within the spatial volume.Advantageously, the magnetic field inside the dish can be deduced from the measured values ​​of the dish based on the field equations for magnetic fields. The arrangement in a dish facilitates the determination of an inverse matrix, described below, for inferring the magnetic resonance signals of the individual samples from the signals of the receiving antennas. In a preferred embodiment, there is at least one distinct axis, for which only one sample is ever arranged in a plane perpendicular to the axis.

[0013] The field camera has N receiving antennas for a magnetic resonance signal. The receiving antennas can, for example, be antenna coils, such as those used in local coils. The receiving antennas each have a receiving volume. The receiving volume is defined as a volume in which the magnetic resonance signal generated by a sample in the receiving antenna is not attenuated by more than dB compared to the maximum level that the sample can generate. With a circular antenna coil, the maximum magnetic resonance signal is achieved, for example, with a sample in the immediate vicinity of the coil conductor.

[0014] The receiving antennas are arranged relative to the spatial volume to be measured such that two of the M samples are located in at least one of the receiving volumes, with the receiving volumes being at least partially disjoint. At least one sample is located in each of the receiving volumes. For example, the samples can be arranged in a grid within a cube or cylinder, on the outside of which the receiving antennas are arranged in all three spatial directions. For example, it would be conceivable to arrange a phantom consisting of a structural element with samples distributed within it in a head coil as a field camera.

[0015] The number N of receiving antennas is greater than or equal to the number M of samples.

[0016] The method according to the invention is used to measure a magnetic field distribution using a magnetic resonance imaging scanner and a field camera. In one step of the method, a sensitivity matrix for the receiving antennas is determined by measuring a sensitivity E mn for each receiving antenna n, with which a signal from the sample m is received by this antenna. This sensitivity can be specified relative to the other antennas or absolutely, for example, in microvolts the amplitude of the received magnetic resonance signal A n of a receiving antenna n for a predetermined excitation of the sample m. The amplitudes are preferably measured using the receivers of the magnetic resonance imaging scanner, which are also used for image reconstruction. However, it would also be conceivable to use separate receivers for this purpose.

[0017] This requires that the signals of the individual samples be separated or differentiated when acquiring the sensitivity matrix. This can be achieved by using a gradient of the magnetic field to separate the frequencies of the individual samples upon receiving the magnetic resonance signal, thus making the signals distinguishable. However, it would also be conceivable to selectively excite only individual samples using a single gradient field when exciting the nuclear spins, provided they are positioned appropriately with respect to the direction of the gradient.

[0018] Finally, in an embodiment which does not fall within the scope of the invention, it would also be possible to specifically achieve selectivity for individual samples by arranging suitable activation devices on the samples. For example, coils on the samples could be used for selection, with which an additional local static magnetic field is generated or the coherent excitation is destroyed by an alternating magnetic field. It would also be conceivable to selectively prevent excitation or the radiation of the MRI signal by switchable shielding on the individual samples. It is also conceivable to determine the sensitivity of the individual receiving antennas by image acquisition of the samples, each with a receiving antenna and the body coil, and to determine the relative sensitivity of the receiving antennas for the individual samples from the ratio of the intensity values ​​for the individual samples.

[0019] Acceleration can be achieved if a complete 3-dimensional acquisition is not performed, but rather one or more projections in two or even just one dimension are acquired from one or more different directions, and their intensity values ​​are evaluated. These projections can be achieved by applying no magnetic field gradient or only a single magnetic field gradient (e.g., Gx or Gy) during the acquisition of the MR signals. This eliminates the need for complete scanning of k-space. This eliminates the need for phase encoding to scan the entire space.

[0020] If the magnetic resonance signals of the individual samples are distinguishable or only one sample is excited at a time, the absolute or relative sensitivity with which the N receiving antennas detect the magnetic resonance signals of sample m can be determined for each sample m. In this way, the coefficients of the sensitivity matrix E mn can be measured in parallel or sequentially.

[0021] In a further step of the method according to the invention, N antenna signals A n of the M samples are detected in a magnetic field to be measured using the N receiving antennas. In contrast to the measurement of the signals for determining the sensitivity matrix E mn, it is no longer necessary for the magnetic resonance signals to differ in frequency or to be detected sequentially. The detection of the antenna signals is again preferably carried out using the receivers of the magnetic resonance imaging scanner.

[0022] In another step of the method according to the invention, M magnetic resonance signals S m of the individual samples are determined from the N antenna signals A n as a function of the sensitivity matrix E m . This involves solving the system of equations A n = E mn x S m for the magnetic resonance signals S m of the samples. Various advantageous methods for solving this equation are listed in the following subclaims and in the description of the figures.

[0023] The frequency of the separated magnetic resonance signals from the samples then provides a measure of the local static magnetic field at the sample location. The result can then be displayed or used to calibrate subsequent measurements.

[0024] The magnetic resonance imaging scanner has a gradient system, and the field camera is arranged within the magnetic resonance imaging scanner. The step of acquiring a sensitivity matrix further includes the step of determining a field gradient that exposes each sample to a different magnetic field. This condition is met if no more than one sample lies on any plane perpendicular to the gradient vector of the field gradient. This determination can, for example, be performed during the design of the field camera by appropriately arranging the samples in space to form a predetermined gradient vector. However, it is also conceivable to determine the gradient vector for a given distribution of the samples. It is also conceivable to determine different gradient vectors.

[0025] In a further step of the method according to the invention, the determined field gradient is output by means of the gradient system during the acquisition of the sensitivity matrix E mn, in particular during the nuclear spin.

[0026] Advantageously, the field camera allows the magnetic resonance signals of the M samples to be reconstructed from the N received signals by the N receiving antennas arranged around the M samples, based on the condition N >= M, and thus to determine the magnetic fields at M positions in the spatial volume to be measured, as explained in more detail in the following procedure. The field camera also allows the use of existing local coils as part of the field camera, thus reducing the effort.

[0027] Advantageously, the inventive method with the field camera allows for parallel and thus rapid imaging with a relatively simple arrangement of samples in a structural body and a local coil matrix, for example, a head coil, without the need for additional equipment. The structural body is preferably a material without any special electrical or magnetic properties of its own, i.e., a non-conductor with a relative dielectric constant and susceptibility constant close to 1 and a Larmor frequency different from that of the samples, for example, Styrofoam.

[0028] Further advantageous embodiments are specified in the subclaims.

[0029] In a non-claimed embodiment of the field camera, the N receiving antennas at least partially surround the external circumference of the spatial volume to be measured. The spatial volume to be measured comprises at least the volume that surrounds the samples, in other words, the contiguous space in which the samples are arranged. For the purposes of the invention, at least partially surrounding the external circumference is considered to mean that the receiving antennas are arranged around the samples in at least three of the six Cartesian spatial coordinates. For example, it is conceivable that the spatial volume describes a cuboid or a sphere that is arranged along the z-axis or the direction of the B0 field of a magnetic resonance imaging scanner, with receiving antennas arranged on the outer walls in the + / - x-direction and + / - y-direction. The samples are arranged therein, for example, in a grid or preferably on one or two concentric spherical shells.Two concentric spherical shells are also conceivable. A cylinder with a coiled antenna array forming its surface would also be possible. For example, receiving antennas could be arranged in all six directions of space on the surfaces of a surrounding prism or cube.

[0030] The outermost arrangement of the receiving antennas advantageously ensures that all samples are detected by the receiving antennas and also weighted differently, so that the sensitivity matrix is ​​not underdetermined.

[0031] In an unclaimed embodiment of the field camera, at least one sample has an inductively coupled first resonant circuit at the Larmor frequency. The term "the sample has a first resonant circuit" means that the first resonant circuit interacts primarily with the one sample on which it is arranged. The signal from the associated sample, for example, generates a resonant signal in the resonant circuit that is more than 6 dB, 12 dB, or 18 dB higher than a signal induced by a neighboring sample with the same excitation. It is essential that the interaction of the resonant circuit with different samples is different so that differentiation is possible using the sensitivity matrix. This can be achieved, for example, by the first resonant circuit having an inductance in the form of a coil that surrounds the sample at a distance that is smaller than a dimension of the sample in a plane in which the coil lies.The Larmor frequency is considered to be the frequency at which the nuclear spins of the sample exhibit a magnetic resonance signal in a static B0 field of a magnetic resonance imaging scanner to be measured, for example, at 1.5 T, 3 T, or 7 T. To achieve resonance, the resonant circuit can have a capacitance in addition to the coil with its own capacitance.

[0032] Advantageously, a first resonant circuit amplifies a signal from the sample by locally increasing the field strength of the excitation pulse as well as resonantly amplifying the magnetic resonance signal upon reception, so that the magnetic field measurement can be performed faster and more reliably.

[0033] In an embodiment not claimed, the field camera has a second resonant circuit with the same resonant frequency as that of the first resonant circuit. The first resonant circuit is inductively coupled to the second resonant circuit, and the second resonant circuit has a larger induction area than the first resonant circuit. Inductive coupling is considered to exist in particular if, with an alternating magnetic field homogeneously penetrating both resonant circuits at the resonant frequency of the two resonant circuits, the amplitude of the currents in the first resonant circuit is 3 dB, 6 dB, 12 dB, or more higher than the amplitude in the first resonant circuit without the second resonant circuit.An identical resonant frequency is considered to be a frequency that generates an amplitude in the respective resonant circuit in the coupled system that is less than 12 dB, 6 dB, or 3 dB lower than an amplitude at the free natural resonant frequency of the resonant circuit with the same alternating field. It is also conceivable for the first resonant circuit to have a different spatial orientation than the second resonant circuit, for example, an angle between a normal vector of an antenna coil of the first resonant circuit and a normal vector of an antenna coil of the second resonant circuit is greater than 10 degrees, 20 degrees, or 30 degrees. It is also conceivable for multiple first resonant circuits to inductively couple multiple samples to a common second resonant circuit.

[0034] Advantageously, the second resonant circuit, due to its larger surface area, can improve the coupling between a transmitting antenna for excitation pulses on one side and / or the receiving antennas on the other side with the samples, allowing for improved temporal resolution through shorter measurement times or spatial resolution through smaller samples. Different orientations can also improve coupling to the excitation pulses, which in the worst case could approach zero if the field vectors of the excitation field are parallel to the surface of the antenna coil of the first resonant circuit.

[0035] In a non-claimed embodiment of the local coil, the resonant circuit comprises a coil and a capacitance, wherein the capacitance is formed from twisted, insulated conductor ends of the coil.

[0036] The varnish-insulated, twisted conductor ends form a small capacitance with a high quality factor for the resonant circuit due to the opposing conductors, which are insulated from each other in the capacitance range. This capacitance does not require additional mechanically vulnerable solder joints or materials that could potentially distort the magnetic resonance measurement.

[0037] It is also conceivable that the resonant circuit has two resonant frequencies, for example by having two oscillating circuits for different resonant frequencies that are coupled to each other.

[0038] Advantageously, the double resonant circuit allows to provide one field camera for two different B0 magnetic field strengths such as 1.5 T and 3 T.

[0039] In one possible embodiment, the method according to the invention comprises the step of determining an inverse matrix I nm to the sensitivity matrix E mn ;

[0040] The term inverse matrix is ​​not limited here to the narrower mathematical term for square matrices with M = N, but also includes, for example, the so-called pseudo-inverse matrices with N > M, for example Moore-Penrose inverse.

[0041] This requires that the system of equations of the sensitivity matrix E mn is not underdetermined. For a square matrix, this is the case if the determinant is non-zero. This can be achieved by appropriately distributing samples and receiving antennas so that non-identical signals or signal combinations of the same samples are detected by two different receiving antennas. This requires that the number N of receiving antennas be at least as large as the number M of samples.

[0042] If the system of equations is overdetermined, i.e., the number of receiving antennas N is greater than the number of samples M, and the arrangement of the samples relative to the receiving coils is such that the same signal combinations are detected by samples in different receiving antennas, a pseudoinverse matrix I nm can be determined from the sensitivity matrix E mn using self-value decomposition or R modification. A solution with a minimal distance and an improved signal-to-noise ratio can be determined using the least squares method.

[0043] The inversion of the matrix can, for example, be carried out in a control unit of the magnetic resonance imaging scanner, but a separate computing unit would also be conceivable.

[0044] In a further possible step of the method according to the invention, the acquired magnetic resonance signals S m of the individual samples are recovered by multiplying the vector AN from the N antenna signals by the inverse matrix I nm. The calculation can also be performed by a control unit of the magnetic resonance imaging scanner or an image reconstruction unit.

[0045] A particular advantage is that the inverse matrix can be repeatedly applied to subsequent measurements of the antenna signals, provided the relative arrangement of the receiving antennas to the samples does not change. This allows for magnetic field measurements and evaluations with a high repetition rate.

[0046] In principle, however, it would also be conceivable to solve the system of equations individually each time. There are numerous mathematical solution methods for such overdetermined systems of equations with erroneous coefficients. It is conceivable that, especially for an overdetermined system of equations, faster and / or more accurate solutions could be found for certain constellations with specific input vectors or only partially populated sensitivity matrices.

[0047] In addition to selectively exciting individual samples, it is also conceivable to obtain a sensitivity matrix by combining multiple projections of the samples along different axes in one or two dimensions. This can be achieved by scanning k-space only along one axis or by recording only the amplitudes.

[0048] In one possible embodiment of the method according to the invention, the step of acquiring a sensitivity matrix further comprises the substep of weighting the antenna signals with a time-dependent window function to sharpen a spectral distribution during the acquisition of the sensitivity matrix in the time domain. For example, a declining exponential function or a Hann function are conceivable.

[0049] The field gradient shifts the magnetic resonance signals of the individual samples to different Larmor frequencies, making the signals distinguishable. However, due to the finite length of the sampling windows and the duration of the detectable magnetic resonance signal, the resolution in Fourier domain may not be sufficient for separation. Advantageously, appropriate weighting with a window function that also takes the exponentially decaying signal strength into account achieves better resolution in the frequency domain.

[0050] The above-described properties, features and advantages of this invention, as well as the manner in which they are achieved, will become clearer and more clearly understood in connection with the following description of the embodiments, which are explained in more detail in connection with the drawings.

[0051] They show: Fig. 1 shows a schematic overview of a magnetic resonance imaging system with a field camera for the method according to the invention; Fig. 2 shows a schematic representation of an arrangement of samples of a field camera for the method according to the invention; Fig. 3 shows a schematic representation of receiving antennas of a field camera for the method according to the invention; Fig. 4 shows a schematic flow chart of a method according to the invention.

[0052] Fig. 1 shows a schematic representation of an embodiment of a magnetic resonance imaging device 1 with a field camera 60 for the method according to the invention.

[0053] The magnet unit 10 has a field magnet 11 that generates a static magnetic field B0 for aligning nuclear spins of samples or the patient in a recording area. The recording area is characterized by an extremely homogeneous static magnetic field B0, whereby the homogeneity particularly affects the magnetic field strength or magnitude. The recording area is almost spherical and arranged in a patient tunnel 16 that extends in a longitudinal direction 2 through the magnet unit 10. A patient bed 30 is movable in the patient tunnel 16 by the traversing unit 36. The field magnet 11 is typically a superconducting magnet that can generate magnetic fields with a magnetic flux density of up to 3T, and even higher in the latest devices. However, permanent magnets or electromagnets with normally conducting coils can also be used for lower field strengths.

[0054] Furthermore, the magnet unit 10 has gradient coils 12, which are designed to superimpose variable magnetic fields in three spatial directions on the magnetic field B0 for spatial differentiation of the acquired imaging regions in the examination volume. The gradient coils 12 are typically coils made of normally conducting wires that can generate mutually orthogonal fields in the examination volume.

[0055] The magnet unit 10 also has a body coil 14 which is designed to radiate a high-frequency signal supplied via a signal line into the examination volume and to receive resonance signals emitted by the patient 100 and to emit them via a signal line.

[0056] A control unit 20 supplies the magnet unit 10 with the various signals for the gradient coils 12 and the body coil 14 and evaluates the received signals.

[0057] Thus, the control unit 20 has a gradient control 21 which is designed to supply the gradient coils 12 with variable currents via supply lines, which provide the desired gradient fields in the examination volume in a time-coordinated manner.

[0058] Furthermore, the control unit 20 has a radio-frequency unit 22, which is designed to generate a radio-frequency pulse with a predetermined temporal profile, amplitude, and spectral power distribution for exciting a magnetic resonance of the nuclear spins in the patient 100. Pulse powers in the kilowatt range can be achieved. The excitation pulses can be transmitted into the patient 100 via the body coil 14 or via a local transmitting antenna.

[0059] A controller 23 communicates via a signal bus 25 with the gradient controller 21 and the high-frequency unit 22.

[0060] On the patient couch 30, a field camera 60 is arranged instead of the patient in order to measure the magnetic field in the patient tunnel 16. The field camera 60 has, as in Fig. 2 und Fig. 3 shown, probes 61 and receiving antennas 62, wherein the receiving antennas 62 are in signal connection with receivers of the radio-frequency unit 22 via a connecting line 33. The signal connection can also be wireless.

[0061] In Fig. 2 the samples 61 of a field camera 60 are shown schematically.

[0062] The field camera 60 has a number of M samples 61, which are preferably distributed on the surface of a spatial volume 70 to be measured. The samples 61 comprise a material active with respect to nuclear magnetic resonance. These can be hydrogen-containing samples 61, such as water or hydrocarbon compounds. Liquid samples 61 can be contained in cuvettes or vials. The samples 61 can be embedded in a matrix or a structural body that does not itself exhibit nuclear magnetic resonance or is active at a different frequency. The size of the samples 61 is a compromise between spatial resolution and sensitivity. The larger the samples 61, the lower the spatial resolution.

[0063] The spatial volume 70 is depicted here as a cube to simplify the representation of the orientation of the spatial axes. However, the spatial volume 70 can also take any other shape; for example, a sphere, ellipsoid, cylinder, prism, or similar arrangements that at least partially fill the recording area are also conceivable. One or more concentric spherical shells on which the samples 61 are arranged are also advantageous, since, due to the magnetic field laws, the fields within a source-free volume can be determined from the fields on the surface.

[0064] For the method according to the invention, it is necessary that the samples 61 are arranged in such a way that only one of them lies on a plane perpendicular to a gradient vector. This makes it possible to Fig. 4 explained method to distinguish the samples 61 based on the Larmor frequency under the influence of the magnetic gradient.

[0065] This can be achieved by arranging the samples 61 in a regular grid whose symmetry axes are suitably tilted relative to the basic axes x, y, and z of the gradient coils 12. However, a suitable random or regular distribution is also conceivable. Finally, it is also conceivable that, although the axes of the sample arrangement 61 are parallel to the axes of the gradient coils 12, a suitable superposition of the magnetic fields of the gradient coils 12 generates a magnetic field gradient just such that the condition is met.

[0066] To accelerate the measurement, it is also conceivable that interference coils 63 are provided around the individual samples 61, which can be supplied with a direct or high-frequency current in order to allow any existing excitation to decay more quickly so that the next measurement can follow more quickly.

[0067] Finally, as explained in connection with the method, the relative sensitivity of the individual receiving antennas 62 for the individual samples 61 can also be determined by means of several magnetic resonance images of the samples 61 with projection onto a two-dimensional surface or a one-dimensional line.

[0068] For the sake of clarity, the receiving antennas 62 of the field camera 61 are shown as an example in a separate Fig. 3 The receiving antennas 62 each have a receiving volume 64 and are arranged relative to the spatial volume to be measured such that two of the M samples are arranged in at least one of the receiving volumes 64, and the receiving volumes 64 are at least partially disjoint, with at least one sample arranged in each receiving volume. The number N of receiving antennas is greater than or equal to the number M of samples.

[0069] In Fig. 3 A possible configuration of receiving antennas 62 of the field camera 60 is shown. The receiving antennas 62 are shown here as antenna coils. The spatial volume 70 is at least partially surrounded by the receiving antennas 62. The samples 61 within the spatial volume 70 are in Fig. 3 For the sake of clarity, they are not shown, but are shown as in Fig. 2 shown arranged within the spatial volume 70.

[0070] In Fig. 3 The receiving antennas 61 are arranged on the outside circumference of a hollow body. As shown, this can also be, for example, a head coil, which, as in a known model with 64 individual receiving coils as receiving antennas 62, receives magnetic resonance signals from the interior, in which the spatial volume 70 with the samples 61 is arranged. Thus, a field camera can be provided with little effort with an existing head coil by using a cost-effective passive matrix with the samples 61.

[0071] The idea of ​​the present invention is that the signals of the individual samples 61 can be recovered from the reception signals of the reception antennas 62 if the signals of the M samples 61 form a solvable linear system of equations with the N antenna reception signals.

[0072] For this purpose, it is necessary that the number N of receiving antennas 62 is greater than or at least equal to the number M of samples. Furthermore, the signals from all samples 61 must be received by at least one receiving antenna 62 each. For example, in Fig. 3 a receiving volume 64 for a receiving antenna 62 is indicated. The contour indicates the area over which signals from samples 61 are attenuated by a predetermined value compared to a sample in the center of the antenna coil; within the scope of the claimed invention, this value is 40 dB.

[0073] The reception volume 64 is limited to the maximum distance at which the reception signal falls below the noise level. The reception volume 64 of the reception antenna 62 is Fig. 3 a club-shaped structure that extends radially inward. The radially outward lobe is not shown for clarity.

[0074] For N=M, the solvability of the system of equations requires that the reception volumes 64 of two receiving antennas 62 are not identical, in the sense that both receiving antennas 62 deliver the same signal levels, or those scaled by the same factor, for the same samples 61. However, a 1:1 assignment of receiving antennas 62 to samples 61 is not required; rather, some receiving antennas 62 can cover a large reception volume 64, while others detect only a single sample 61.

[0075] To determine the linear system of equations between samples 61 and signals from the receiving antennas 62, one way is to determine the received signal of all N receiving antennas 62 for each individual sample 61, one after the other, for all M samples 61. This is possible, for example, if only one sample 61 is excited or the signals are distinguishable.

[0076] According to the invention, a magnetic field gradient is applied, in which each sample 61 is exposed to a different static magnetic field consisting of magnetic field B0 and gradient field. Then, either a single sample 61 can be excited with a narrowband excitation pulse, and the individual signal of this sample 61 can be evaluated by the receiving antennas 62.

[0077] Or, with broadband excitation, the signals from all samples 61 could be received with all receiving antennas 62 and separated in the frequency domain based on the different frequencies. Measures for improved frequency separation are described below for the method according to the invention.

[0078] In an embodiment not claimed, however, it would also be conceivable, in an embodiment with interference coils 63 around the individual samples, to selectively excite one sample at a time with an excitation pulse via this interference coil 63.

[0079] An exemplary head coil with 64 receiving antennas 62 could therefore, in conjunction with up to 64 samples 61, measure a magnetic field inside the head coil. To do so, it is merely necessary to arrange the samples 61 in a matrix or molded body as already described and then position this in a predetermined position in the head coil so that, as already described, individual excitation and different received signals are possible. It is also conceivable to use resonant coils on the samples as a first resonant circuit to improve the coupling of the samples 61 to the excitation pulse and / or the receiving antennas 62 and thus improve the signal strength and the SNR. The first resonant circuit can also have two different resonant frequencies, e.g., by coupling two resonant circuits, so that signal amplification is possible even with different static B0 fields.

[0080] A two-stage concept with a first resonance circuit directly at the sample and a second resonance circuit with a greater distance, but also a larger induction area of ​​the antenna coil, could also further improve the coupling.

[0081] In Fig. 4 a schematic flow chart of a method according to the invention for measuring a magnetic field distribution with a magnetic resonance imaging device 1 and a field camera 60 is given.

[0082] In a step S100, a sensitivity matrix for the receiving antennas with a sensitivity E mn for each sample m is acquired at each receiving antenna n. For this purpose, it is necessary to acquire the signal response for at least M, or better yet N with N > M, receiving antennas 62 for a predetermined excitation of each sample m. For a signal vector with the magnetic resonance signals S m of the M samples 61 and an antenna signal vector with the received signals A n of the N receiving antennas 62, the system of equations results: A n = E mn × S m

[0083] One possible embodiment for determining the matrix elements of the sensitivity matrix is ​​to excite each of the M samples 61 individually one after the other. This can be achieved, for example, by generating a magnetic field gradient in a step S120, in which each of the M samples 61 is exposed to a magnetic field of different magnitude. This can be achieved, as already described, by arranging the samples relative to the magnetic field gradient such that only one sample 61 is arranged in a plane perpendicular to the gradient vector. The samples 61 are either aligned in the X, Y, or Z direction according to a gradient vector of the individual gradient coils 12, so that it is sufficient to apply current to the respective gradient coil 12.Or, in a step S110, a suitable gradient vector is determined and, in a step S120, a gradient vector in a suitable direction relative to the samples 61 is generated by superimposing the fields of the three gradient coils 12.

[0084] If each of the M samples is exposed to a different magnetic field composed of the gradient magnetic field and the static magnetic field B0, they also exhibit different Larmor frequencies. A narrowband excitation pulse with the respective Larmor frequency can then be used to individually excite a single sample, and the magnetic resonance signal of this single sample 61 can then be received by all N receiving antennas 62 to detect the respective matrix elements of the sensitivity matrix E mn.

[0085] It would also be conceivable to emit a broadband excitation pulse under the influence of the same magnetic field gradient, which includes signal components with the Larmor frequency of all samples under the influence of this magnetic field gradient and the static magnetic field B0. The M samples are then all excited simultaneously. If the magnetic resonance signals are subsequently recorded with the receiving antennas 62 with unchanged magnetic fields, they represent a superposition of several magnetic resonance signals from different samples 61. However, since the Larmor frequencies differ, a separation in the frequency domain can be performed, for example, after a Fourier transformation, and thus the individual elements of the sensitivity matrix E mn can be determined.

[0086] If the frequency spacing is insufficient for clean separation, it is also conceivable to increase the selectivity by weighting the received signals of the receiving antennas 62 with a window function such as the Hann or Gaussian function. Furthermore, exponentially increasing weighting can compensate for the natural decay of the magnetic resonance signal and reduce the linewidth.

[0087] It would also be conceivable to increase the selectivity by combining selective excitation with frequency-domain analysis. Either individual samples could be selectively excited or several spaced-apart slices could be used.

[0088] Finally, in an embodiment not claimed, it would also be conceivable, in an embodiment with a disturbance coil 63 around the individual samples 61, to excite the disturbance coils 63 for a selective local excitation of an individual sample 61 by means of an excitation pulse applied to the respective disturbance coil 63.

[0089] In a step S310, an inverse matrix I nm is determined for the sensitivity matrix E mn . This requires that the number N of receiving antennas be at least as large as the number M of samples. The sensitivity matrix E mn specifies a system of equations that is reversible under certain conditions, so that the magnetic resonance signals of the individual samples 61 can be inferred from the received signals or the amplitude A n of the receiving antennas 62.

[0090] This is done mathematically using a matrix I nm which is inverse to E mn. A square matrix E nm is invertible if the determinate is not equal to 0. Then the vector magnetic resonance signals results from the received signals A n of the receiving antennas 62 according to S m = I nm x A n.

[0091] If the number N of receiving antennas 62 is greater than the number M of samples 61, the term inverse matrix is ​​not limited to the narrower mathematical term for square matrices with M = N, but also includes the so-called pseudo-inverse matrices with N > M, such as Moore-Penrose inverse.

[0092] This requires that the system of equations of the sensitivity matrix E mn is not underdetermined. If the system of equations is overdetermined, i.e., the number N of receiving antennas 62 is greater than the number M of samples 61 and the arrangement of the samples 61 relative to the receiving antennas 62 is such that non-identical linear combinations of signals from samples 61 are detected in different receiving antennas 62, a pseudoinverse matrix I nm can be determined from the sensitivity matrix using self-value decomposition or R modification. From this, in turn, a solution with a minimal distance and improved signal-to-noise ratio for the magnetic resonance signals S m of the samples 61 can be determined using the least squares method.

[0093] The inversion of the matrix can be carried out, for example, in a control unit 20 of the magnetic resonance imaging device 1, but a separate computing unit or a computing unit used for image reconstruction would also be conceivable.

[0094] In a further step S200, N antenna signals A n of the M samples 61 are acquired in a magnetic field to be measured using the N receiving antennas 62. This first requires excitation of the samples 61 using an excitation pulse with the Larmor frequency at an expected magnetic field B0. Since the separation of the magnetic resonance signals of the individual samples 61 is then carried out based on the spatial diversity of the sensitivity matrix E mn or its inverse I nm, it is not necessary to expose the individual samples 61 to a different magnetic field using a gradient. The excitation can therefore be carried out using a narrow-band common excitation pulse whose frequency is the Larmor frequency of the samples 61 at the expected static magnetic field B0. The bandwidth of the excitation pulse only needs to be large enough to compensate for inhomogeneities in the static magnetic field B0. The receiving antennas 62 then acquire or measure the N amplitude values.the vector of antenna signals A n of the M samples 61.

[0095] In a step S320, the magnetic resonance signals S m of the individual samples 62 are then determined by multiplying the vector AN by the inverse matrix I nm. For an overdetermined matrix with N > M, a pseudo-inverse matrix or a system of equations, which can be solved, for example, using the least squares method, is used instead of the quadratic inverse matrix.

[0096] Advantageously, the inverse matrix remains unchanged as long as the sensitivity matrix is ​​not altered, for example, by spatial changes. This allows the inverse matrix to be used for rapid calculation of the magnetic resonance signals for a large number of rapidly consecutive measurements.

[0097] However, it is also conceivable that, under special configurations of receiving antennas 62 and / or samples 61, other solution methods for the underlying equation system A n = E mn x S m may provide a more accurate or faster result. It is also conceivable that the solution of the equation method is performed individually after each measurement depending on the respective antenna signals and adapted accordingly.

[0098] An evaluation of the individual magnetic resonance signals S m of the individual samples 61 obtained in this way according to frequency, for example by a Fourier transformation, in turn results in a value that is directly proportional to the magnetic field at the location of the sample 61. The evaluation is preferably carried out on the control unit 20 of the magnetic resonance tomograph 1

[0099] The magnetic field value thus determined can be output to a user via a display to evaluate the homogeneity of the magnetic field B0. However, it is also conceivable that the control unit 20 determines a setting for shim currents through shimming coils of the magnetic resonance imaging device 1 from the B0 values ​​and outputs this setting to improve the homogeneity of the static magnetic field B0.

[0100] Although the invention has been illustrated and described in detail by the preferred embodiment, the invention is not limited by the disclosed examples and other variations may be derived therefrom by those skilled in the art without departing from the scope of the invention, which is defined solely by the claims.

Claims

1. Method for measuring a magnetic field distribution in a spatial volume to be measured, using a magnetic resonance tomograph (1) and a field camera (60) for capturing a magnetic field distribution by means of a magnetic resonance measurement, wherein the field camera (60) has: a number of M samples (61), which are distributed over a spatial volume to be measured (70); a number of N receive antennas (62), wherein the receive antennas (62) each have a receive volume (64), in which the magnetic resonance signal produced in the receive antenna by the sample is not attenuated by more than 40 dB relative to a maximum level that can be generated by the sample, and are arranged relative to the spatial volume to be measured (70) in such a way that two of the M samples (61) are arranged in at least one of the receive volumes (64) and the receive volumes (64) are at least partially separate, and at least one sample (61) is arranged in each receive volume (64), wherein the number N of receive antennas (62) is greater than or equal to the number M of samples (61), wherein the method has steps as follows: (S100) capturing a sensitivity matrix for the receive antennas (62), with a sensitivity Emn for each sample (61) m at each receive antenna (62) n, using the magnetic resonance tomograph (1); (S200) capturing, by means of the N receive antennas (62), N antenna signals An of the M samples (61) in a magnetic field to be measured, using the magnetic resonance tomograph (1); (S300) determining the M magnetic resonance signals Sm of the individual samples from the N antenna signals An as a function of the sensitivity matrix Emn, using a controller (23), wherein the magnetic resonance tomograph (1) has a gradient system and the field camera (60) is arranged in the magnetic resonance tomograph (1), characterised in that the step (S100) of capturing a sensitivity matrix additionally has the substeps: (S110) determining a magnetic field gradient, under whose influence each sample (61) is subjected to a different magnetic field; (S120) generating the determined field gradient by means of the gradient system during the capture of the sensitivity matrix Emn.

2. Method according to claim 1, wherein the method additionally has the steps: (S310) defining the inverse matrix Inm to the sensitivity matrix Emn; (S320) ascertaining the magnetic resonance signals Sm of the individual samples by multiplying the vector AN from the N antenna signals with the inverse matrix Inm.

3. Method according to claim 1, wherein the step (S100) of capturing a sensitivity matrix additionally has the substeps: (S130) weighting the antenna signals with a time-dependent window function in order to sharpen a spectral distribution during the capture of the sensitivity matrix.

4. Method according to one of the preceding claims, wherein the N receive antennas (62) of the field camera (60) at least partially surround the outer extent of the spatial volume (70).

5. Method according to one of the preceding claims, wherein at least one sample (61) of the field camera (60) has a first resonant circuit with a resonance at the Larmor frequency, wherein the field camera has a second resonant circuit, wherein the first resonant circuit is inductively coupled to the second resonant circuit, and the second resonant circuit has a larger induction surface than the first resonant circuit.

6. Method according to claim 5, wherein the first resonant circuit has a coil and a capacitor, wherein the capacitor is formed from twisted insulated conductor ends of the coil.

7. Method according to claim 5 or 6, wherein the first resonant circuit has two different resonant frequencies.

8. Computer program product which can be loaded directly into a processor of a programmable controller (23) of a magnetic resonance tomograph (1) with a field camera (60) according to claim 1, with program code means for executing all steps of a method according to one of claims 1 to 7 when the program product is executed on the controller (23).

9. Computer-readable storage medium on which is stored electronically readable control information that is so configured as to perform the method according to one of claims 1 to 7 when the storage medium is used in a controller (23) of a magnetic resonance tomograph (1) with a field camera (60) according to claim 1.