Magnetic resonance method for reducing residual magnetization and adjusting the amplitude of gradient pulses

[0002] The present invention relates to an MR (magnetic resonance) imaging method and MRI (magnetic resonance imaging) apparatus, and more particularly to an MR imaging method and MRI apparatus which can reduce residual magnetization caused by gradient pulses.

[0003] In prior art, Japanese Patent Application Laid Open No. H10-75940 discloses a technique involving:

[0004] (1) a phase shift measurement method comprising the steps of: executing a prescan sequence comprising transmitting an excitation RF pulse, transmitting an inversion RF pulse, applying a phase encoding pulse on a phase gradient axis, applying a read pulse on a read gradient axis and applying a rewinder pulse on the phase gradient axis, and subsequently transmitting an inversion RF pulse, applying a dephaser pulse on the phase gradient axis and collecting data from an echo while applying a read pulse on the phase gradient axis; and measuring a phase shift in subsequent echoes caused by the effect of eddy current or residual magnetization from the phase encoding pulse and so forth based on phase data obtained by performing one-dimensional Fourier transformation on the collected data, and

[0005] (2) an MR imaging method employing a fast spin echo pulse sequence that involves repeatedly executing the following steps a plurality of times with a varying phase encoding pulse: after transmitting an excitation RF pulse, transmitting an inversion RF pulse, applying a phase encoding pulse on a phase gradient axis, collecting data from an echo while applying a read pulse on a read gradient axis, and applying a rewinder pulse on the phase gradient axis, thereby collecting data for a plurality of echoes with one excitation RF pulse, wherein a compensation pulse for compensating for a phase shift amount measured by the phase shift measurement method as described regarding (1) is either incorporated in the phase encoding pulse or appended either immediately before or immediately after or immediately before and after the phase encoding pulse, or incorporated in the rewinder pulse or appended either immediately before or immediately after or immediately before and after the rewinder pulse.

[0006] The above technique presupposes that the phase shift amount measured by the phase shift measurement method of (1) is equal to the phase shift amount generated in the fast spin echo pulse sequence of (2) before the compensation pulse is added.

[0007] However, these amounts are not always equal in the MRI apparatus because of the magnetic hysteresis characteristics of a magnetism conditioning plate or the like, and the presupposition in the above technique does not always hold. This will be explained with reference to FIGS. 1 and 2 hereinbelow.

[0008] FIG. 1 is a pulse sequence chart according to a conventional fast spin echo (FSE) technique.

[0009] In an FSE sequence SQ, an excitation RF pulse R and a slice selective pulse ss are first applied. Next, a dephasing pulse gx1 is applied on a read gradient axis. Next, a first inversion RF pulse P1 and a slice selective pulse ss are applied. Next, a phase encoding pulse gy1i is applied on a phase gradient axis. Then, data is collected from a first echo echo1 while applying a read pulse gxw. Thereafter, a rewinder pulse gy1ri is applied on the phase gradient axis having the same area as, and a polarity opposite to, the phase encoding pulse gy1i. Reference symbol i represents a repetition number of the FSE sequence SQ in FIG. 1, and i=1—I (for example, I=128).

[0010] Next, a second inversion RF pulse P2 and a slice selective pulse ss are applied, a phase encoding pulse gy2i is applied on the phase gradient axis, data is collected from a second echo echo2 while applying a read pulse gxw, and then, a rewinder pulse gy2ri is applied on the phase gradient axis having the same area as, and a polarity opposite to, the phase encoding pulse gy2i.

[0011] Thereafter, and similarly, a j-th inversion RF pulse Pj and a slice selective pulse ss are applied, a phase encoding pulse gyji is applied on the phase gradient axis, data is collected from a j-th echo echoj while applying a read pulse gxw, and thereafter, a rewinder pulse gyjri is applied on the phase gradient axis having the same area as, and a polarity opposite to, the phase encoding pulse gyji, repeatedly for j=3—J (although J=8 for example, FIG. 1 shows a case that J=3).

[0012] And finally, a killer pulse of a large amplitude is applied on the phase gradient axis.

[0013] FIG. 2 is diagram of the magnetic hysteresis characteristics of a ferromagnetic material such as a magnetism conditioning plate in the MRI apparatus.

[0014] The magnetization strength B of the ferromagnetic material such as the magnetism conditioning plate varies as indicated by a main loop Ma when the external magnetic field strength H is substantially changed, while it varies as indicated by a minor loop Mi when the change in the external magnetic field strength H is small. A gradient pulse corresponds to the small change in the external magnetic field strength H. Accordingly, application of a gradient pulse causes the magnetic strength B of the ferromagnetic material such as the magnetism conditioning plate to vary as indicated by the minor loop Mi.

[0015] Thus, the MRI apparatus has residual magnetization varying depending on a history of applying gradient pulses, owing to the magnetic hysteresis characteristics of a ferromagnetic material such as a magnetism conditioning plate.

[0016] However, since the prescan sequence as described regarding (1) does not take care of residual magnetization due to the killer pulse kp, the phase shift amount measured by the phase shift measurement method of (1) is not equal to the phase shift amount generated in the fast spin echo pulse sequence of (2) before the compensation pulse is added. That is, residual magnetization due to a killer pulse in an (i−1)th FSE sequence SQ affects all the echoes in an i-th FSE sequence SQ.

[0017] Moreover, the prescan sequence of (1) is in the form of a partially cut-out FSE sequence up to the first echo, and the history of applying gradient pulses of the prescan sequence is not equal to that of the MR imaging scan at and after the second echo. Therefore, the residual magnetization affects the second echo and the following echoes.

[0018] Thus, the conventional technique as described above has a problem that the effect of residual magnetization caused by gradient pulses cannot be sufficiently reduced.

[0019] It is therefore an object of the present invention to provide an MR imaging method and MRI apparatus which can sufficiently reduce the effect of residual magnetization due to gradient pulses.

[0020] In accordance with a first aspect of the present invention, there is provided an MR imaging method comprising the steps of applying a gradient pulse having either a positive or negative polarity on a gradient axis, and thereafter applying a residual magnetization reducing pulse having a polarity and amplitude to reduce residual magnetization caused by the gradient pulse.

[0021] In the MR imaging method of the first aspect, a residual magnetization reducing pulse is applied after a gradient pulse is applied. The residual magnetization reducing pulse has a polarity opposite to the gradient pulse, and has an amplitude that can reduce residual magnetization caused by the gradient pulse. Thus, the residual magnetization after the application of the residual magnetization reducing pulse is reduced to a negligible degree. Therefore, the effect of residual magnetization caused by a gradient pulse can be sufficiently reduced.

[0022] In accordance with a second aspect of the invention, there is provided an MR imaging method comprising the steps of applying a killer pulse on a gradient axis, and thereafter applying a residual magnetization reducing pulse having a polarity and amplitude to reduce residual magnetization caused by the killer pulse.

[0023] In this configuration, the killer pulse is a gradient pulse for eliminating transverse magnetization by forcible dephasing. The killer pulse is also referred to as a spoiler pulse.

[0024] In the MR imaging method of the second aspect, a residual magnetization reducing pulse is applied after a killer pulse is applied. The residual magnetization reducing pulse has a polarity opposite to the killer pulse, and has an amplitude that can reduce residual magnetization caused by the killer pulse. Thus, the residual magnetization after the application of the residual magnetization reducing pulse is reduced to a negligible degree. Therefore, residual magnetization due to a killer pulse kp in an (i−1)th FSE sequence SQ can be prevented from affecting data of an i-th FSE sequence SQ.

[0025] In accordance with a third aspect of the invention, there is provided an MR imaging method in which a phase encoding pulse is applied on a phase gradient axis, the method comprising the steps of: applying a residual magnetization reducing pulse after the phase encoding pulse, the residual magnetization reducing pulse having a polarity and amplitude to reduce residual magnetization caused by the phase encoding pulse, and additively increasing the area of the phase encoding pulse by the area of the residual magnetization reducing pulse.

[0026] In the MR imaging method of the third aspect, a residual magnetization reducing pulse is applied after a phase encoding pulse is applied. The residual magnetization reducing pulse has a polarity opposite to the phase encoding pulse, and has an amplitude that can reduce residual magnetization caused by the phase encoding pulse. Thus, the residual magnetization after the application of the residual magnetization reducing pulse is reduced to a negligible degree. Therefore, residual magnetization due to a phase encoding pulse can be prevented from affecting the subsequent data.

[0027] In accordance with a fourth aspect of the invention, there is provided an MR imaging method in which a phase encoding pulse is applied on a phase gradient axis and a rewinder pulse is applied after collecting an NMR signal, the method comprising the steps of: applying a residual magnetization reducing pulse after the phase encoding pulse, the residual magnetization reducing pulse having a polarity and amplitude to reduce residual magnetization caused by the phase encoding pulse, and additively increasing the area of the phase encoding pulse by the area of the residual magnetization reducing pulse, and further, applying a residual magnetization reducing pulse after the rewinder pulse, the residual magnetization reducing pulse having a polarity and amplitude to reduce residual magnetization caused by the rewinder pulse, and additively increasing the area of the rewinder pulse by the area of the residual magnetization reducing pulse.

[0028] In the MR imaging method of the fourth aspect, a residual magnetization reducing pulse is applied after a phase encoding pulse is applied. The residual magnetization reducing pulse has a polarity opposite to the phase encoding pulse, and has an amplitude that can reduce residual magnetization caused by the phase encoding pulse. Moreover, a residual magnetization reducing pulse is applied after a rewinder pulse is applied. The residual magnetization reducing pulse has a polarity opposite to the rewinder pulse, and has an amplitude that can reduce residual magnetization caused by the rewinder pulse. Thus, the residual magnetization after the application of the residual magnetization reducing pulse is reduced to a negligible degree. Therefore, residual magnetization due to a phase encoding pulse and a rewinder pulse can be prevented from affecting the subsequent data.

[0029] In accordance with a fifth aspect of the invention, there is provided an MR imaging method in which two or more gradient pulses having different polarities are successively applied on a gradient axis, comprising the step of determining respective amplitudes of the two or more gradient pulses to reduce residual magnetization after successively applying the two or more gradient pulses.

[0030] In the MR imaging method of the fifth aspect, because two or more gradient pulses having different polarities are successively applied, any new residual magnetization reducing pulse is not added, but instead, respective amplitudes of the two or more gradient pulses are adjusted to reduce residual magnetization to a negligible degree after successively applying the two or more gradient pulses. Therefore, residual magnetization after successively applying two or more gradient pulses can be prevented from affecting data.

[0031] In accordance with a sixth aspect of the invention, there is provided an MR imaging method in which a slice selective pulse is applied on a slice gradient axis and subsequently a rephasing pulse is applied, the method comprising the step of determining an amplitude of the rephasing pulse to reduce residual magnetization caused by the slice selective pulse.

[0032] In the MR imaging method of the sixth aspect, because a slice selective pulse and a rephasing pulse having different properties are successively applied, the amplitude of the rephasing pulse is adjusted to reduce residual magnetization to a negligible degree after successively applying the slice selective and rephasing pulses. Therefore, residual magnetization after successively applying a slice selective pulse and a rephasing pulse can be prevented from affecting data.

[0033] In accordance with a seventh aspect of the invention, there is provided an MR imaging method in which a dephasing pulse is applied on a read gradient axis and thereafter a read pulse is applied, comprising the step of determining an amplitude of the dephasing pulse so that the read pulse can reduce residual magnetization caused by the dephasing pulse.

[0034] In the MR imaging method of the seventh aspect, because a dephasing pulse and a read pulse having different polarities are successively applied, the amplitude of the dephasing pulse is adjusted to reduce residual magnetization to a negligible degree after successively applying the dephasing and read pulses. Therefore, residual magnetization after successively applying a dephasing pulse and a read pulse can be prevented from affecting data.

[0035] In accordance with an eighth aspect of the invention, there is provided an MR imaging method in which gradient moment nulling (GMN) phase compensation pulses are applied on a gradient axis, the method comprising the step of determining respective amplitudes of the GMN phase compensation pulses to reduce residual magnetization after applying the GMN phase compensation pulses.

[0036] In the above configuration, the GMN phase compensation pulse is defined as a gradient pulse having a waveform designed to eliminate motion-induced phase variation of nuclear spins.

[0037] In the MR imaging method of the eighth aspect, because GMN phase compensation pulses in which two or more gradient pulses having different polarities are combined are applied, the respective amplitudes of the gradient pulses are adjusted to reduce residual magnetization to a negligible degree after successively applying the GMN phase compensation pulses. Therefore, residual magnetization after successively applying GMN phase compensation pulses can be prevented from affecting data.

[0038] Thus, according to the MR imaging method and MRI apparatus of the present invention, for a gradient pulse having either a positive or negative polarity, a residual magnetization reducing pulse is applied after applying the gradient pulse, or for two or more successively applied gradient pulses having different polarities, the amplitude of the gradient pulse(s) is adjusted to reduce residual magnetization thereafter, and hence, an unwanted phase error does not occur and image quality degradation such as ghosting or shading can be prevented.

[0039] Further objects and advantages of the present invention will be apparent from the following description of the preferred embodiments of the invention as illustrated in the accompanying drawings.

[0040] FIG. 1 is a chart of an exemplary pulse sequence of a conventional FSE sequence.

[0041] FIG. 2 is a diagram for explaining magnetic hysteresis characteristics.

[0042] FIG. 3 is a block diagram illustrating an MRI apparatus in accordance with a first embodiment of the present invention.

[0043] FIG. 4 is a chart of an exemplary pulse sequence of an FSE sequence in accordance with the first embodiment.

[0044] FIG. 5 is a diagram for explaining.a fourth embodiment in which the present invention is applied to a combination of a slice selective pulse and a rephasing pulse.

[0045] FIG. 6 is a diagram for explaining a fifth embodiment in which the present invention is applied to a combination of a dephasing pulse and a read pulse.

[0046] FIG. 7 a diagram for explaining a sixth embodiment in which the present invention is applied to phase compensation pulses.

[0047] The present invention will now be described in more detail with reference to several embodiments shown in the accompanying drawings.

[0048] First Embodiment

[0049] FIG. 3 is a block diagram of an MRI apparatus in accordance with a first embodiment of the present invention.

[0050] In the MRI apparatus 100, a magnet assembly 1 has therein an empty portion (bore) for inserting a subject, and surrounding the empty portion are disposed a permanent magnet 1pfor applying a static magnetic field having a constant strength H₀to the subject, a gradient magnetic field coil 1gfor applying gradient pulses on a slice gradient axis, phase gradient axis and read gradient axis, a transmit coil 1tfor supplying an RF pulse for exciting spins of atomic nuclei within the subject, and a receive coil 1rfor detecting an NMR signal from the subject. The gradient magnetic field coil 1g,transmit coil 1t and receive coil 1rare connected to a gradient magnetic field drive circuit 3, an RF power amplifier 4 and a preamplifier 5, respectively.

[0051] It should be noted that a superconductive or normal conductive magnet may be employed instead of the permanent magnet.

[0052] A sequence memory circuit 8 operates the gradient magnetic field drive circuit 3 following instructions from a computer 7 based on a stored pulse sequence, to drive the gradient magnetic field coil 1gin the magnet assembly 1 to apply a gradient pulse. The sequence memory circuit 8 also operates a gate modulation circuit 9 to modulate a carrier output signal from an RF oscillation circuit 10 into a pulsed signal having a certain timing and envelope shape. The pulsed signal is applied as an RF pulse to the RF power amplifier 4, power-amplified in the RF power amplifier 4, and applied to the transmit coil 1tin the magnet assembly 1 to selectively excite a desired slice region.

[0053] The preamplifier 5 amplifies an NMR signal detected by the receive coil 1rin the magnet assembly 1 from the subject, and inputs the signal to a phase detector 12. The phase detector 12 phase-detects the NMR signal from the preamplifier 5 with reference to the carrier output signal from the RF oscillation circuit 10, and supplies the phase-detected signal to an A/D converter 11. The A/D converter 11 converts the phase-detected analog signal into a digital signal, and inputs it to the computer 7.

[0054] The computer 7 reads the data from the A/D converter 11, and performs image reconstruction operation to produce an image of the desired slice region. The image is displayed on a display device 6. The computer 7 is also responsible for overall control such as receiving information input from an operator console 13.

[0055] FIG. 4 is a chart of a pulse sequence of a fast spin echo technique in accordance with the present invention.

[0056] In the FSE sequence SQ, an excitation RF pulse R and a slice selective pulse ss are first applied. Next, a dephasing pulse gx1 is applied on a read gradient axis. Next, a first inversion RF pulse P1 and a slice selective pulse ss are applied.

[0057] Next, a phase encoding pulse gy1i is applied on a phase gradient axis. Subsequently, a residual magnetization reducing pulse gy1i_rms is applied having an opposite polarity and an amplitude to cancel residual magnetization due to the phase encoding pulse gy1i. It should be noted that the area of the phase encoding pulse gy1i is increased by the area of the residual magnetization reducing pulse gy1i_rms to prevent the phase encoding amount from being varied by the residual magnetization reducing pulse gy1i_rms.

[0058] The amplitude that can cancel the residual magnetization due to the phase encoding pulse gy1i is basically about half the amplitude of the phase encoding pulse gy1i because the magnetic hysteresis characteristic curve has a point symmetrical shape. However, since the amplitude may be varied depending on material of a magnetism conditioning plate or characteristics of the gradient magnetic field coil 1g,it is preferred that the amplitude be determined by simulation using a mathematical model of the minor loop Mi or by experiment.

[0059] Then, data is collected from a first echo echol while applying a read pulse gxw.

[0060] Thereafter, a rewinder pulse gy1ri is applied on the phase gradient axis having the same area as, and a polarity opposite to, the phase encoding pulse gy1i. Subsequently, a residual magnetization reducing pulse gy1ri_rms is applied having an opposite polarity and an amplitude to cancel residual magnetization due to the rewinder pulse gy1ri. It should be noted that the area of the rewinder pulse gy1ri is increased by the area of the residual magnetization reducing pulse gy1ri_rms to prevent the phase encoding amount from being varied by the residual magnetization reducing pulse gy1ri_rms. Although the amplitude of the residual magnetization reducing pulse gy1ri_rms is basically about half the amplitude of the rewinder pulse gy1ri similarly to above, it is preferred that the amplitude be determined by simulation or by experiment.

[0061] Reference numeral i is the repetition number of the FSE sequence SQ shown in FIG. 4, and i=1—I (I=128, for example).

[0062] Then, a second inversion RF pulse P2 and a slice selective pulse ss are applied, a phase encoding pulse gy2i is applied on the phase gradient axis, and a residual magnetization reducing pulse gy2i_rms is applied similarly to above. Next, data is collected from a second echo echo2 while applying a read pulse gxw. Thereafter, a rewinder pulse gy2ri is applied on the phase gradient axis having the same area as, and a polarity opposite to, the phase encoding pulse gy1i, and a residual magnetization reducing pulse gy2ri_rms is applied similarly to above.

[0063] Thereafter, and similarly, a j-th inversion RF pulse Pj and a slice selective pulse ss are applied, a phase encoding pulse gyji and a residual magnetization reducing pulse gyji_rms are applied on the phase gradient axis, data is collected from a j-th echo echoj while applying a read pulse gxw, and then a rewinder pulse gyjri having the same area as, and a polarity opposite to, the phase encoding pulse gyji, and a residual magnetization reducing pulse gyjri_rms are applied on the phase gradient axis, repeatedly for j=3—J (although J=8 for example, FIG. 4 shows a case that J=3).

[0064] And finally, a killer pulse kp of a large amplitude is applied on the phase gradient axis, and subsequently a residual magnetization reducing pulse kp_rms is applied having an opposite polarity and an amplitude to cancel residual magnetization due to the killer pulse kp.

[0065] According to the MRI apparatus 100, residual magnetization due to a phase encoding pulse, rewinder pulse and killer pulse can be reduced, and therefore image quality degradation such as ghosting and shading can be restrained from being generated in an image by the residual magnetization.

[0066] Second Embodiment

[0067] Subsequent to a killer pulse of an inversion recovery technique or a killer pulse for a saturation pulse, a residual magnetization reducing pulse may be applied having an opposite polarity and an amplitude to cancel residual magnetization due to such killer pulses.

[0068] Third Embodiment

[0069] A residual magnetization saturation pulse of a maximum amplitude may be applied immediately before an excitation pulse, and subsequently a residual magnetization reducing pulse may be applied having an opposite polarity and an amplitude to cancel residual magnetization due to the residual magnetization saturation pulse.

[0070] Fourth Embodiment

[0071] In a pulse sequence in which a slice selective pulse ss is followed by a rephasing pulse sr having an opposite polarity, as shown in FIG. 5, the amplitude of the rephasing pulse sr may be adjusted to reduce residual magnetization due to the slice selective pulse ss. However, the time width of the rephasing pulse sr should be also adjusted in order not to vary the area.

[0072] This can reduce residual magnetization after applying the slice selective pulse ss followed by the rephasing pulse sr.

[0073] Fifth Embodiment

[0074] In a gradient echo-type pulse sequence in which an opposite-polarity dephasing pulse dp is applied before a read pulse ro as shown in FIG. 6, the amplitude of the dephasing pulse dp may be adjusted to reduce residual magnetization due to the read pulse ro. However, the time width of the dephasing pulse dp should be also adjusted in order not to vary the area.

[0075] This can reduce residual magnetization after applying the read pulse ro followed by the dephasing pulse dp.

[0076] Sixth Embodiment

[0077] In a pulse sequence in which first-order GMN phase compensation pulses are applied having three combined gradient pulses fc1, fc2 and fc3 of different polarities, as shown in FIG. 7, The respective amplitudes of the gradient pulses fc1, fc2 and fc3 may be adjusted so that residual magnetization after applying the phase compensation pulses becomes as small as possible. However, the respective time widths of the gradient pulses fc1, fc2 and fc3 should also adjusted in order not to vary the area.

[0078] This can reduce residual magnetization after applying phase compensation pulses.

[0079] Many widely different embodiments of the invention may be configured without departing from the spirit and the scope of the present invention. It should be understood that the present invention is not limited to the specific embodiments described in the specification, except as defined in the appended claims.