Method and apparatus for magnetic resonance imaging on the basis of a gradient echo sequence

[0001] 1. Field of the Invention

[0002] The present invention generally concerns magnetic resonance tomography (MRT) as used in medicine for examination of patients. The present invention more particularly concerns a magnetic resonance tomography apparatus as well as a method for operation of such a magnetic resonance tomography apparatus, with which a high contrast can be achieved for imaging on the basis of a gradient echo sequence.

[0003] 2. Description of the Prior Art

[0004] MRT is based on the physical phenomenon of nuclear magnetic resonance and has been successfully used for over 15 years as an imaging modality in medicine and in biophysics. In this examination modality the subject is exposed to a strong, constant magnetic field. The nuclear spins in the atoms in the subject, which were previously randomly oriented, are aligned.

[0005] Radio-frequency energy can now excite these “ordered” nuclear spins to a specific oscillation. This oscillation generates the actual measurement signal in MRT, the measurement signal being acquired by suitable acquisition coils. The measurement subject can be spatially coded in all three spatial directions by the use of non-homogeneous magnetic fields generated by gradient coils. The method allows a free selection of the slice to be imaged, so slice images of the human body can be acquired in all directions. MRT as a slice imaging method in medical diagnostics is distinguished as a non-invasive examination method primarily due to its versatile contrast capability. MRT has developed into a method superior to x-ray computed tomography (CT) due to the exceptional display capability of soft tissue. MRT today is based on the use of spin echo and gradient echo sequences that enable an excellent image quality with measurement times on the order of minutes.

[0006] The continuous technical development of the components of MRT apparatuses and the introduction of faster imaging sequences opens MRT to ever more fields of use in medicine. Real-time imaging to support minimally-invasive surgery, functional imaging in neurology and perfusion measurement in cardiology are only a few examples. In spite of the technical progress in the design of MRT apparatuses, image contrast and signal-noise ratio (SNR) of the MRT image remain limiting factors for many applications of MRT in medical diagnostics.

[0007] Particularly in the case of image acquisitions of the head, the goal is to ensure a good segmentation, meaning a good contrast between grey brain matter, white brain matter and cerebrospinal fluid (CSF). One possibility for this is a (semi)automated method for segmentation using T1-weighted MPRAGE data. In the following the MPRAGE sequence as well as the associated fundamentals are therefore initially described.

[0008] The acquisition of the data in MRT occurs in k-space (frequency domain). The MRT image in the image domain is linked with the MRT data in k-space by a Fourier transformation. The spatial coding of the subject which spans k-space occurs by means of gradients in all three spatial directions. For this purpose, auxiliary magnetic fields Gx, Gy and Gz, whose field strengths linearly depend on the respective spatial coordinates x, y and z, are superimposed on the homogeneous basic magnetic field. Without limitation as to generality, in the further discussion Cartesian k-space is assumed that is scanned (sampled) per slice or per line.

[0009] In MRT imaging the gradient fields are used in different ways. In selective slice excitation a gradient field is superimposed on the homogeneous basic field along one of the coordinate axes (typically the z-axis) during the RF pulse. By selection of a specific frequency spectrum of the RF pulse, only nuclei within a specific slice perpendicular to the z-axis are excited. For frequency coding, a gradient field (typically along the x-axis) is superimposed on the basic magnetic field during the acquisition of the RF signal. The readout of the RF signal ensues in N-equidistant time steps Δt. For phase coding a gradient field (typically along the y-axis) with a constant gradient strength is superimposed on the basic magnetic field for a specific time ty before the acquisition of the RF signal. The readout ensues by repetition of the sequence N times, with the gradient strength being increased in equidistant steps per repetition.

[0010] FIG. 2A shows the principle of MRT imaging with the 2D Fourier method. A slice is established by selective slice excitation along the z-axis; this slice of k-space is subsequently scanned line-by-line. The data of a single k-space line are frequency-coded by means of a gradient Gy upon readout. The readout of a line ensues in N equidistant time steps. Each line in k-space has the interval Δkx that is generated by a phase coding step. The imaging sequence is repeated N times for various values of the phase coding gradient Gx. In total, a number matrix with N×N data points is obtained, from which an MRT in the image domain can be constructed by 2D Fourier transformation. FIG. 2B shows the 3D Fourier method. The slice-selection gradient is replaced by a second phase coding gradient. This means that the entire volume of the nuclei is excited by the RF pulse and the spatial information is coded exclusively by orthogonal gradients, namely by two phase coding gradients and one frequency coding gradient. M slices perpendicular to the z-axis are acquired, each slice being scanned line-by-line in k-space. The coding within a slice of k-space ensues by a frequency coding gradient in the y-direction as well as a phase coding gradient in the x-direction. A number matrix with M×N×N data packets is thus obtained in total.

[0011] FIG. 3 schematically shows the excitation and gradient scheme of the known FLASH (Fast Angle Low Shot) sequence. This is based on the principle of the gradient echo technique. Fast image sequences that are based on the principle of small angle excitation, and in which the echo signals generated exclusively by gradient reversal, are designated as gradient echo sequences (GE sequences). In small angle excitation, flip angles of á<90° are used, but only a small fraction of the longitudinal magnetization is rotated in the transversal plane. Thus it is not necessary to wait as long for the relaxation of the magnetization, which leads to significant time savings. Furthermore, the dephasing of the transverse magnetization caused by the two gradients is compensated by the polarity reversal, such that a gradient echo arises. In FIG. 3 the RF pulse with a small angle excitation below an angle is shown in the first line and the RF signal with the gradient echo is subsequently shown on the time axis. In the second line the slice-selection gradient Gz is plotted along the time. As already explained, the slice-selection gradient is superimposed on the homogeneous magnetic field along the z-axis during the RF pulse and the slice-selection gradient is subsequently reversed in terms of polarity for the purpose of dephasing. In the third line the frequency coding gradient Gy is shown along the time axis. A gradient field in the y-direction is superimposed on the homogeneous magnetic field for the frequency coding after polarity reversal of the gradient during the acquisition of the RF signal. The phase coding gradient Gx is shown along the time axis in line 4. For phase coding along the x-axis, a constant gradient is hereby switched on for a defined time before acquisition of the RF signal and the sequence is repeated Nx times. The transverse magnetization is destroyed after the data acquisition via spoiler gradients switched in each of the three spatial axes after acquisition of the RE signal. The echo time TE designated in FIG. 3 is the time from the radiation of the RF pulse up to the gradient echo and the repetition time TR is the time for a sequence pass.

[0012] The scheme of the k-space scan of a gradient echo sequence is shown in FIG. 4. After the RF pulse the signal is located in the center of k-space (1). A dephasing of the signal at the point (2) occurs due to the phase coding gradients and the dephasing in the readout direction, A k-space line is scanned (3, 4) during the reverse-polarized readout gradients and the signal is acquired. The gradient echo occurs at the point (3). The entire process is repeated Nx times with phase coding gradients of various strengths such that an image of the entirety of k-space is generated.

[0013] FIG. 5 shows the basic principle of the MPRAGE (Magnetization Prepared Rapid Gradient Echo) sequence. This is based on the 3D Fourier method as well as the magnetization preparation. A preparation phase is activated before the actual image phase to achieve shorter measurement times and a good tissue contrast. The preparation phase effects a preparation of the magnetization that is dependent on the relaxation times T1 and T2. The magnetization prepared in this manner is spatially coded and scanned using the gradient echo sequence. FIG. 5 schematically shows the workflow of the MPRAGE sequence in which a magnetization preparation initially occurs; in the imaging phase all Fourier lines are subsequently acquired in the x-direction given a constant value kz along the z-axis. A recovery phase follows for a better SNR and thus a better contrast, and the sequence is subsequently repeated for further values of kz.

[0014] The MPRAGE sequence is used by default for depiction of T1-weighted images of the head with good contrast of the grey and white brain matters. The images are used both for routine clinical examinations, but also increasingly in recent times for automatic determination/segmentation of volumes of the brain, specific brain regions or specific tissue types (morphometry). A requirement for this purpose is that the data produce a good contrast between white brain matter and grey brain matter and a good contrast between cerebrospinal fluid (CSF) and grey brain matter.

[0015] In (semi-)automated methods for segmentation using T1-weighted MPRAGE data, the inversion time TI (i.e. the time from the beginning of the sequence up to reaching the k-space center during the scan of the gradient echo sequence) is typically set such that a compromise between grey-white contrast and grey-CSF contrast is made. In case of doubt, a manual segmentation or description of the contrast limits is then required. Moreover, the consequence is that an error in the description of the contrast limits is tolerated. This is a problem particularly at high field strengths, since the B1 homogeneity is generally poorer and makes it difficult to achieve completely automatic segmentation, in addition to causing dielectric resonance effects.

[0016] An object of the present invention is to provide a method and an apparatus for MRT imaging in which a good contrast is ensured both between grey and white brain matters and between grey brain matter and CSF.

[0017] This object is achieved according to the invention by a method for magnetic resonance imaging on the basis of a gradient echo sequence by excitation of nuclear spins in a subject and acquisition of radio-frequency signals caused by the excited nuclear spins, whereby including the following steps:

[0018] a) preparation of the magnetization of the spins by an inversion pulse;

[0019] b) implementation of a number of steps for spin excitation as well as acquisition of an RF response signal for a first image contrast, with the measurement data being acquired along a first two-dimensional slice, and this first two-dimensional slice being parallel to a plane spanned by two coordinate axes x, y that are orthogonal to one another;

[0020] c) implementation of a number of steps of the spin excitation, as well as acquisition of an RF response signal for a second image contrast, with the measurement data being acquired along the first two-dimensional slice that was present in step b); and

[0021] d) repetition of the steps a) through c) for further two-dimensional slices that are offset parallel to the first two-dimensional slice along a third coordinate axis z that is orthogonal to the first two coordinate axes x and y.

[0022] The above object also is achieved in accordance with the invention by an apparatus for magnetic resonance imaging on the basis of a gradient echo sequence by excitation of nuclear spins and acquisition of radio-frequency signals arising from the excited nuclear spins with: a device for preparation of the magnetization of the spins by means of an inversion pulse; a device for implementation of a number of steps for spin excitation as well as measurement of an RF response signal for a first image contrast, wherein the measurement data are acquired along a first two-dimensional slice, and wherein this first two-dimensional slice is parallel to a plane spanned by two coordinate axes x, y that are orthogonal to one another; a device for implementation of a number of steps for spin excitation as well as acquisition of an RF response signal for a second image contrast, wherein the measurement data are acquired along the first two-dimensional slice; and a device for repetition of all steps for further two-dimensional slices that are offset parallel to the first two-dimensional slice along a third coordinate axis z that is orthogonal to the first two coordinate axes x and y.

[0023] The position of the two-dimensional slices along the z-coordinate axis is advantageously established by phase coding.

[0024] Furthermore, the measurement data are advantageously established along the two-dimensional slice in the direction of the x-coordinate axis by phase coding.

[0025] In an embodiment the measurement data along the two-dimensional slice in the direction of the y-coordinate axis are established by frequency coding.

[0026] A wait time can be provided after the inversion pulse.

[0027] The acquisition of the first image contrast and the second image contrast can be done with the following steps:

[0028] I) excitation of the spins by irradiation of an RF pulse,

[0029] II) acquisition of measurement data along the two-dimensional slice by readout of all values along the y-coordinate axis for a fixed value of the x-coordinate axis, and

[0030] III) repetition of the steps I) and II) for various values along the x-coordinate axis.

[0031] A weighting of the longitudinal relaxation time T1 can ensue.

[0032] The time TI1 from the inversion pulse up to the irradiation of the RF pulse is advantageously selected for the first image contrast such that the contrast between grey brain matter and white brain matter is particularly high for the first image contrast.

[0033] Furthermore, the time TI2 from the inversion pulse up to the irradiation of the RF pulse for the second image contrast is advantageously selected such that the contrast between grey brain matter and cerebrospinal fluid is particularly high for the second image contrast.

[0034] The time TR between two inversion pulses can be between 2000 ms and 2600 ms.

[0043] FIG. 1 is a schematic representation of a magnetic resonance imaging (magnetic resonance tomography) apparatus for generation of a magnetic resonance image of a subject according to the present invention. The design of the magnetic resonance tomography apparatus corresponds to the design of a conventional tomography apparatus, with the differences noted below. A basic field magnet 1 generates a temporally constant strong magnetic field for polarization or alignment of the nuclear spins in the examination region of the subject (such as, for example, a part of the human body to be examined). The high homogeneity of the basic magnetic field that is required for the magnetic resonance measurement is defined in a spherical measurement volume M into which the parts of the human body to be examined are introduced. Shim plates made from ferromagnetic material are mounted at suitable points to support the homogeneity requirements and in particular to eliminate temporally-invariable influences. Temporally-variable influences are eliminated by shim coils 2 that are activated by a shim power supply 15.

[0044] A cylindrical gradient coil system 3 that has three sub-windings is located in the basic field magnet 1. Each sub-winding is supplied with current from an amplifier 14 for generation of a linear gradient field in the respective directions of the Cartesian coordinate system. The first sub-coil of the gradient field system 3 generates a gradient G_x in the x-direction, the second sub-coil generates a gradient G_y in the y-direction and the third sub-coil generates a gradient G_z in the z-direction. Each amplifier 14 has a digital-analog converter that is activated by a sequence controller 18 for time-accurate generation of the gradient pulses.

[0045] Located within the gradient field system 3 is a radio-frequency antenna 4 that converts the radio-frequency pulses emitted by a radio-frequency power amplifier 16 into an alternating magnetic field for excitation of the nuclei and alignment of the nuclear spins of the subject to be examined or of the region of the subject to be examined. The radio-frequency antenna 4 has one or more RF transmission coils and a number of RF reception coils in the form of an advantageously linear arrangement of component coils. The alternating field originating from the precessing nuclear spins (i.e. normally the nuclear spin echo signals caused by a pulse sequence made up of one or more radio-frequency pulses and one or more gradient pulses) is also converted by the RF reception coils of the radio-frequency antenna 4 into a voltage that is supplied via an amplifier 7 to a radio-frequency reception channel 8 of a radio-frequency system 22. The radio-frequency system 22 furthermore has a transmission channel 9 in which the radio-frequency pulses are generated for the excitation of the nuclear magnetic resonance. The respective radio-frequency pulses are digitally represented in the sequence controller 18 as a series of complex numbers based on a pulse sequence predetermined by the system computer 20. This number series is supplied as a real part and an imaginary part to respective inputs 12 of a digital-analog converter in the radio-frequency system 22, and from this to a transmission channel 9. In the transmission channel 9 the pulse sequences are modulated on a radio-frequency carrier signal having a base frequency that corresponds to the resonance frequency of the nuclear spins in the measurement volume.

[0046] The switch-over from transmission mode to reception mode ensues via a transmission-reception diplexer 6. The RF transmission coil of the radio-frequency antenna 4 radiates the radio-frequency pulses for excitation of the nuclear spins into the measurement volume M for excitation of the nuclear spins and samples resulting echo signals via the RF reception coils. The acquired magnetic resonance signals are phase-sensitively demodulated on an intermediate frequency in the reception channel 8 of the radio-frequency system 22, and converted into a real part and an imaginary part of the measurement signal in the analog-digital converter. An image is reconstructed by an image computer 17 from the measurement data acquired in this manner. The administration of the measurement data, the image data and the control programs ensues via the system computer 20. Based on control programs, the sequence controller 18 monitors the generation of the respective desired pulse sequences and the corresponding scanning of k-space. The sequence controller 18 controls the time-accurate switching of the gradients, the emission of the radio-frequency pulses with defined phase and amplitude and the reception of the nuclear magnetic resonance signals. The time base for the radio-frequency system 22 and the sequence controller 18 is provided by a synthesizer 19. The selection of corresponding control programs for generation of a nuclear magnetic resonance image as well as the representation of the generated nuclear magnetic resonance image ensues via a terminal 21 that has a keyboard as well as one or more screens.

[0047] FIG. 2 schematically shows the workflow of the sequence according to the present invention The sequence is based on the MPRAGE sequence. As already explained, in the MPRAGE sequence a recovery time is inserted after the acquisition of the RF signal. In this recovery time only a minimal contrast of grey brain matter and white brain matter is still present, but the contrast between grey brain matter and CSF is particularly good during this recovery time. According to the present invention, a further data set is acquired during this recovery time, which acquisition can possibly occur with higher bandwidth in order to conclude the measurement during the recovery time. The standard measurement can occur at an earlier point in time, which can possibly achieve an even better contrast between grey brain matter and white brain matter.

[0048] The homogeneous basic magnetic field is generated by means of the basic field magnet 1. According to the present sequence a magnetization preparation is thus implemented in step S1 by an inversion pulse. Here the inversion pulse can advantageously amount to 180°. In a next step S2 a wait time is inserted in order to obtain a better contrast or a better SNR. In the next step S3 an RF pulse is radiated by the RF antenna 4, ideally with a flip angle á of <90°. The imaging for the first contrast, i.e. for the contrast between grey brain matter and white brain matter, ensues by readout of the RF signal in the step S4 in k-space all k_y are read out for a fixed k_x. According to step S5 the steps S3 and S4 are repeated for various values of k_x. In step S6 an RF pulse (likewise with a flip angle á of <90°) is subsequently newly radiated by means of the RF antenna 4. The imaging for the second contrast, i.e. for the contrast between grey brain matter and CSF, ensues in the subsequent step S7. For this purpose all k_y for a respective fixed k_x are read out again. In step S8 the steps S6 and S7 are repeated for various values of k_x. Finally, in step S9 all steps of the sequence S1 through S7 are repeated for various values of k_z. The coding along the gradients G_x, G_y and G_z ensues with the gradient coil system 3 which is controlled by the sequence controller 18.

[0049] FIG. 7 shows an exemplary embodiment according to the sequence of the present invention. The time axis is plotted to the right, the pulse scheme is shown in the first line and the gradient schematics are shown in the further three lines. The inversion pulse for magnetization preparation ensues at the beginning of the sequence. After the first inversion time TI₁ the RF pulse is radiated by the RF antenna 4. At the same time the gradient fields G_z, G_y and G_x are subsequently switched on by the sequence controller 18. The gradient G_z allows a phase coding in the direction of the z-coordinate axis, the gradient G_x allows a phase coding along the x-coordinate axis, and the gradient G_y (due to activation and subsequent reverse-polarization) causes a cancellation of the dephasing of the transverse magnetization, whereby the gradient echo is induced. In the subsequent acquisition phase k-space is spatially coded in the direction of the y-axis by the gradient G_y. The workflow of the sequence from radiation of the RF pulse up to the termination of the readout of the first image contrast is ultimately repeated for various values k_x of the gradient G_x. After a further time TI2 the radiation of the second RF pulse ensues by means of the RF antenna 4 for the second image contrast. In correspondence with the acquisition of the first image contrast, gradients G_z and G_x for phase coding in the x-direction and z-direction are activated by the sequence controller 18 along with a gradient G_y for spatial coding along the y-coordinate axis. The sequence for the second image contrast is likewise repeated for various values of k_x. The time TD designates the recovery phase in the original MPRAGE sequence which, according to the present invention, is now used for the imaging of the second contrast. The repetition time TR shows the time for the passage of a sequence, i.e. the time from irradiation of an inversion pulse up to the irradiation of a next inversion pulse. The sequence is repeated for various values of k_z.

[0050] A reliable demarcation of grey brain matter, white brain matter and CSF is enabled by the acquisition of a further data set with different contrast. This can alternatively be realized by separate evaluation or by calculated/combined images from the two (or even more) contrasts by processing of the measurement data in the image computer 17.

[0051] It is advantageous that the use of two or more contrasts increases the precision of segmentation methods. This moves a step closer to a fully-automated evaluation of such data sets, since manual contributions are presently still required. It is also advantageous that the total measurement time normally does not need to be increased. If the various contrasts are already combined with a suitable algorithm during the image recon, the work expenditure of the user is not altered by additional images.

[0052] Although modifications and changes may be suggested by those skilled in the art, it is the intention of the inventor to embody within the patent warranted hereon all changes and modifications as reasonably and properly come within the scope of his contribution to the art.