JP5214209B2 - Magnetic resonance imaging system - Google Patents

Description

  The present invention relates to a magnetic resonance imaging apparatus that separates and images various substances related to metabolism.

  In addition to magnetic resonance imaging (MRI), which is currently widely used, as an imaging method using a magnetic resonance imaging apparatus, a clue is made to the difference in resonance frequency (chemical shift) due to the difference in chemical bonds of various molecules including hydrogen nuclei. In addition, there is a method (magnetic resonance spectroscopic imaging (MRSI)) for separating and imaging magnetic resonance signals for each molecule.

  By using this MRSI, the distribution of metabolites inside the human body can be imaged non-invasively, but the concentration of metabolites contained in the subject is usually very low. On the other hand, the concentration of water contained in the human body is very high, so a process that suppresses the generation of water signals is added immediately before normal excitation and detection to suppress unnecessary water signals and necessary metabolites. Can be detected (water suppression measurement).

  In addition, since the magnitude of the chemical shift is very small on the order of ppm, in MRSI, it is very important to adjust the magnetic field uniformity that affects the magnetic resonance frequency. In general, since the influence of the subject on the magnetic field uniformity is quite large, it is necessary to improve the static magnetic field uniformity with the subject placed in the static magnetic field. Therefore, a method of adjusting the magnetic field uniformity by changing each offset value of three different gradient magnetic fields and the amount of current flowing through each shim coil and superimposing the magnetic fields generated by each gradient coil and each shim coil on the static magnetic field. Has been proposed (see Non-Patent Document 1).

  In this method, a combination of current-magnetic field distribution characteristics of each gradient magnetic field coil and each shim coil obtained from the reference image (a current value passed through each coil) is calculated so that the magnetic field distribution in the subject becomes uniform. As the reference image and the target image used here, an MRI phase distribution image is generally used. Normally, when measuring this phase distribution image, “a modified spin echo method in which the spin echo time and the gradient echo time are shifted by Δt” is used (for example, see Patent Document 1).

  However, the static magnetic field distribution at the time of signal measurement by the modified spin echo method and the static magnetic field distribution at the time of signal measurement by the MRSI measurement sequence do not necessarily match. This is because when a gradient magnetic field is applied, a large eddy current is generated on the inner surface of the bore or the subject surface in the magnet, and the magnitude and time constant of the magnetic field caused by this eddy current (change in the magnitude of the magnetic field over time). This is because the gradient magnetic field changes depending on the application intensity, application time, and signal detection time, and therefore the degree of magnetic field inhomogeneity caused by eddy currents differs depending on the sequence shape.

  In particular, the signal detection time of the modified spin echo method is usually as short as several milliseconds to tens of milliseconds, while the signal detection time of the MRSI measurement sequence is as long as several tens of milliseconds to 1 second. It is very difficult to correct a long eddy current by the shimming.

  Therefore, in MRSI measurement, in order to improve measurement accuracy, distortion correction of measurement signals caused by this eddy current is performed as data post-processing. In order to cancel the influence of the eddy current, a phase compensation circuit that shapes the gradient magnetic field waveform, a gradient magnetic field with an active shield, and an eddy current correction function that cancels the crosstalk component between the gradient magnetic field axes are used.

  As an example of an eddy current correction function that cancels the crosstalk component, there is a technique described in Patent Document 2. In this known technique, the temporal change of the gradient magnetic field is measured using a spectrum measurement signal, but the magnetic field change characteristic measured using a specific sequence is used as a correction table for performing crosstalk compensation between axes. When the gradient magnetic field is stored and output in each other normal sequence, the corrected gradient magnetic field waveform is calculated with reference to the correction table.

  Next, an eddy current correction method using data post-processing will be described (for example, see Non-Patent Document 2). In this method, first, in the measurement of the reference image, in order to detect the magnetic field inhomogeneity caused by the eddy current with high accuracy, image data that has been excited and detected without suppressing the water signal is used (non-MRSI). Water suppression measurement).

  Next, water suppression measurement is performed to obtain the spectrum of the metabolite. Then, a correction function that minimizes the influence of the eddy current on the water signal peak is derived for each spectrum of each spatial point for the obtained non-water suppression measurement data. Finally, by applying the derived correction function to the water suppression measurement data for each same spatial point, it becomes possible to reduce the influence of eddy current on the signal peak of each metabolite. .

Japanese Patent Publication No. 05-056140 Japanese Patent Laid-Open No. 08-089494 Journal of Magnetic Resonance, Vol. 77, pp. 40-52 (1988) Magnetic Resonance in Medicine, Vol. 14, pp. 26-30 (1990)

  As a fundamental problem in the prior art, when water suppression measurement is performed in a situation where magnetic field inhomogeneities caused by eddy currents remain, the water signal suppression rate and SNR are naturally reduced. Rather than performing data correction later, it is desirable to reduce the effect of eddy currents that accompany the measurement sequence before measurement.

  However, with the eddy current correction function in the prior art, it is difficult to completely cancel the eddy current on the inner surface of the magnet bore. Further, there is no effect on the eddy current on the surface of the subject, and it is difficult to say that a necessary and sufficient reduction effect is obtained.

  In addition, in the eddy current correction method using data post-processing, if the non-water suppression measurement performed to acquire the reference image is performed under the same measurement conditions as the water suppression measurement that is the main measurement, the reference measurement time is included. The total measurement time is doubled, and the measurement time is too long to be applied to clinical diagnosis.

  As a method for shortening the time required for the reference measurement, a method for reducing the number of repetitions of measurement has been proposed as the most effective method. However, since information obtained with a decrease in the number of measurements is also reduced (a fine space) Information is lost), and the actual eddy current correction effect is also reduced.

  An object of the present invention is to realize a magnetic resonance imaging apparatus capable of reducing magnetic field inhomogeneity due to eddy currents and obtaining a highly accurate magnetic resonance spectrum image.

  In order to achieve the above object, the present invention is configured as follows.

  In the magnetic resonance imaging apparatus, the time variation of the static magnetic field inhomogeneity is calculated using the nuclear magnetic resonance signal measured without suppressing the nuclear magnetic resonance signal from water, and the calculated time variation of the static magnetic field inhomogeneity is calculated. A static magnetic field adjustment amount for canceling is calculated, and the static magnetic field adjustment means is controlled using the calculated static magnetic field adjustment amount.

  Since this measurement, which suppresses nuclear magnetic resonance signals from water, is performed with reduced static magnetic field inhomogeneity, high-accuracy images can be acquired without requiring data post-processing. it can.

  That is, a magnetic resonance imaging apparatus capable of reducing magnetic field inhomogeneity due to eddy currents and obtaining a highly accurate magnetic resonance spectrum image can be realized.

  Embodiments of the present invention will be described below with reference to the accompanying drawings.

  1A to 1C are external views of an MRI apparatus to which the present invention is applied. 1A is a horizontal magnetic field type MRSI apparatus using a tunnel-type magnet that generates a static magnetic field with a solenoid coil, and FIG. 1B is a hamburger in which the magnets are separated into upper and lower parts to enhance the feeling of opening. This is a type (open type) vertical magnetic field type MRI apparatus.

  1 (c) is a tunnel-type MRI apparatus, similar to the apparatus shown in FIG. 1 (a), but it increases the feeling of opening by shortening the depth of the magnet and tilting it obliquely. This is an MRI apparatus. The present invention can be applied to MRI apparatuses having other structures including these MRSI apparatuses.

  FIG. 2 is a schematic block diagram of an MRI apparatus to which the present invention is applied. This MRI apparatus includes a static magnetic field coil 2 for generating a static magnetic field, a gradient magnetic field coil 3 for applying gradient magnetic fields in three directions orthogonal to each other, and a high-frequency magnetic field for the subject 1. And a receiving high-frequency coil 6 for receiving a magnetic resonance signal generated from the subject 1. In some cases, a shim coil 4 that can adjust the uniformity of the static magnetic field is provided.

  Various types of static magnetic field coils 2 are employed depending on the structure of the apparatus shown in FIG. The gradient magnetic field coil 3 and the shim coil 4 are driven by a gradient magnetic field power supply unit 12 and a shim power supply unit 13, respectively. The high frequency magnetic field irradiated by the transmission coil 5 is generated by the transmitter 7 and applied to the subject 1 placed in a static magnetic field.

  The example shown in FIG. 2 shows a configuration in which the transmission coil 5 and the reception coil 6 are separately provided, but there is also a configuration in which only one high-frequency coil that is used for both transmission and reception is used.

  The magnetic resonance signal detected by the receiving coil 6 is sent to the computer 9 through the receiver 8. The computer 9 performs various arithmetic processing on the magnetic resonance signal according to a program registered in advance or an instruction from the user, and generates spectrum information and image information. In the MRI apparatus according to the embodiment of the present invention, the calculator 9 calculates the magnetic resonance signal detected by the receiving coil 6. The computer 9 is connected to a display 10, a storage device 11, a sequence control device 14, an input device 15, and the like. The generated spectrum information and image information described above are displayed on the display 10 and recorded in the storage device 11. To do. The input device 15 is used to input measurement conditions, conditions necessary for arithmetic processing, and the like, and these measurement conditions and the like are recorded in the storage device 11 as necessary.

  The sequence control device 14 controls the drive power supply unit 12 for the gradient magnetic field generating coil 3, the drive power supply unit 13 for the shim coil 4, the transmitter 7, and the receiver 9 in accordance with a program registered in advance or an instruction from the user. Control. The control time chart (pulse sequence) is set in advance by the imaging method and is stored in the storage device 11. In the MRI apparatus according to the embodiment of the present invention, the sequence controller 14 suppresses the nuclear magnetic resonance signal from the pulse sequence that performs MRI measurement (non-water suppression measurement) without suppressing the nuclear magnetic resonance signal from water. And a pulse sequence for performing MRSI measurement (water suppression measurement), and executing these two kinds of pulse sequences in combination. As these pulse sequences, known pulse sequences can be employed. An example is shown in FIGS.

  FIG. 3 is a diagram illustrating an example of an MRSI pulse sequence. In one embodiment of the invention, this pulse sequence is performed for both non-water suppression measurements and water suppression measurements. FIG. 4 is a diagram showing an example of a pre-pulse sequence executed before water suppression measurement.

  In accordance with the MRSI pulse sequence shown in FIG. 3, the sequence controller 14 performs the following control. First, a first gradient magnetic field (gradient magnetic field in the X-axis direction) Gs1 for selecting the first slab (planar region perpendicular to the X-axis) and a first high-frequency magnetic field RF1 called a 90 ° pulse are simultaneously applied. The nuclear magnetization in the first slab is brought into an excited state. Here, TE is an echo time and TR is a repetition time.

  Next, after the elapse of TE / 4 from the irradiation of the high-frequency magnetic field RF1, the second gradient magnetic field (gradient magnetic field in the Y-axis direction) Gs2 for selecting the second slab (planar region perpendicular to the Y-axis), the 180 ° pulse, A second high-frequency magnetic field RF2 called is applied at the same time, and the nuclear magnetization contained in the second slab among the nuclear magnetization in the first slab excited by RF1 is reversed by 180 °.

  Further, after TE / 2 has elapsed from the irradiation of RF2, a third gradient magnetic field (gradient magnetic field in the Z-axis direction) Gs3 for selecting a third slab (planar region perpendicular to the Z-axis) and a third pulse called a 180 ° pulse are used. The high-frequency magnetic field RF3 is simultaneously applied, and the nuclear magnetization included in the third slab among the nuclear magnetization in the intersecting region of the first slab and the second slab reversed by RF2 is reversed 180 ° again.

  By applying the above-described three sets of high-frequency magnetic field and gradient magnetic field, a magnetic resonance signal Sig1 is generated in which the time after TE / 4 elapses from the irradiation of RF3 is the eco-time. The generated magnetic resonance signal Sig1 is measured for j points at a predetermined sampling interval.

  Then, the application of the gradient magnetic fields Gp1, Gp2, and Gp3 for phase encoding that can give three-dimensional spatial information is changed stepwise, and the measurement of Sig1 is repeated. Since the magnetic resonance signal Sig1 having the signal change in the time axis direction includes the information on the chemical shift described above, a magnetic resonance spectrum signal is obtained by performing Fourier transform in the time axis direction as described later. be able to.

  Gs1 'applied immediately after application of the gradient magnetic field Gs1 is a gradient magnetic field for rephasing (phase return) with respect to Gs1. In addition, Gd1 and Gd1 ′, Gd2 and Gd2 ′, and Gd3 and Gd3 ′ applied before and after the application of RF2 do not disturb the phase of the nuclear magnetization excited by the irradiation of RF1, but are excited by the irradiation of RF2. This is a gradient magnetic field for dephasing the phase. Further, Gd4 and Gd4 ′, Gd5 and Gd5 ′, and Gd6 and Gd6 ′ applied before and after the application of RF3 do not disturb the phase of the nuclear magnetization excited by the irradiation of RF1, and the nuclear magnets excited by the irradiation of RF3. This is a gradient magnetic field for dephasing the phase.

  By executing the pulse sequence of FIG. 3 described above, it is possible to selectively excite only the nuclear magnetization included in the region where the three slabs intersect (excitation region: R1). Then, the application intensity of the phase encoding gradient magnetic fields Gp1, Gp2 and Gp3 is changed, Gp1 is in the a stage, Gp2 is in the b stage, and Gp3 is in the c stage. A magnetic resonance signal is detected, and a series of data filling a measurement space (kx, ky, kz) called a k space (a space related to the real space through Fourier transformation) is measured.

  In addition, kx means the x-axis component of k space, and a means the integer below A showing the measurement number of a kx-axis direction. Similarly, ky and kz represent the y-axis and z-axis components of the k space, respectively, and b and c represent integers equal to or less than B and C representing the measurement numbers in the ky axis and the kz direction, respectively. In the case of a two-dimensional k-space, any one of A, B, and C is 1, and the other two are integers of 2 or more. In the case of a three-dimensional k-space, all of A, B, and C are two or more. It will be an integer.

  Next, the obtained a × b × c magnetic resonance signals are subjected to inverse Fourier space for converting the k space component and the real space component in each k space axis (kx, ky, kz) direction. The magnetic resonance spectrum generated from each of the a × b × c space points can be obtained by performing Fourier transform for converting the time component and the frequency component in the time axis direction.

  On the other hand, when water suppression measurement is performed, the sequence control device 14 performs control according to the prepulse sequence for suppressing the water signal shown in FIG. 4 immediately before performing excitation / detection according to the MRSI sequence of FIG. . In other words, in the water suppression measurement, a pulse sequence that is a set of the pulse sequence shown in FIG. 4 and the MRSI sequence shown in FIG. 3 is executed.

  In the prepulse sequence shown in FIG. 4, first, in order to excite nuclear magnetization contained only in water molecules, the transmission frequency Ft is set to the resonance frequency Fw of water, and the excitation frequency band ΔFt is set to the water peak width. Irradiation with a high frequency magnetic field (high frequency magnetic field for water excitation) RFw1 set to about ΔFw is performed (selective excitation of water nuclear magnetization).

  Next, the phase gradient magnetic field Gdw1 is applied in order to separate the phase of nuclear magnetization of water in an excited state and make the vector sum of water nuclear magnetization zero (pseudo saturation of water nuclear magnetization). . Further, in order to increase the suppression effect of the water signal, the application of the high frequency magnetic field and the phase gradient magnetic field similar to the water excitation high frequency magnetic field RFw1 and the phase gradient magnetic field Gdw1 is repeated a plurality of times. FIG. 4 shows an example of a sequence that is repeated three times, but the number of repetitions is not limited to this. If it is not necessary to distinguish between them, hereinafter, the high-frequency magnetic field is represented by RFw and the phase gradient magnetic field is represented by Gdw1.

  In many cases, a Gaussian waveform having a narrow-band excitation frequency characteristic is used for the high-frequency magnetic field RFw. The example shown in FIG. 4 is an example in which a gradient magnetic field of any one of Gx, Gy, and Gz is applied as the gradient magnetic field for phase. However, the gradient magnetic fields of all three axes Gx, Gy, and Gz may be applied simultaneously, or any two axes may be applied simultaneously. Then, while the pseudo saturation state of the water magnetization continues, the MRSI sequence shown in FIG. 3 is performed to measure a weak metabolite signal. Usually, the flip angle of the high frequency magnetic field RFw for water excitation is often set to around 90 °, but various combinations and numerical values can be used as the number of applied axes and the applied intensity for the gradient magnetic field Gdw for phase.

  After the flip angle of all the water excitation high-frequency magnetic fields RFw shown in FIG. 4 is set to 0 °, the non-water suppression measurement is performed by executing the MRSI sequence shown in FIG. It can also be done. In this case, the influence of the eddy current caused by the gradient magnetic field Gdw for phase at the time of non-water suppression measurement can be made the same as that at the time of water suppression measurement.

  That is, both the non-water suppression measurement and the water suppression measurement are configured so that the prepulse sequence shown in FIG. 4 is performed before the MRSI sequence shown in FIG. 3, and the flip angles of all the water excitation high-frequency magnetic fields RFw shown in FIG. Can be switched between non-water suppression measurement and water suppression measurement.

  In addition, since metabolite signals that can be detected from within the living body are often very weak, measurement under the same conditions (the phase encoding gradient magnetic field is not changed) for the purpose of improving the SNR of the spectrum obtained. (Measurement) may be repeated a plurality of times, and a process of adding the obtained signals may be performed (integration process).

  Next, an eddy current correction method in MRSI according to an embodiment of the present invention will be described. Note that the reference measurement refers to measuring a reference image indicating current-magnetic field distribution characteristics of each gradient magnetic field coil and each shim coil with a simulated sample as a measurement target in advance.

  FIG. 5 is a diagram showing an outline of a measurement procedure in one embodiment of the present invention. In FIG. 5, first, by executing the above-described non-water suppression measurement (reference measurement), a two-dimensional k-space or a three-dimensional k-space (kx (a), ky (b), kz (c)) Spectrum information at the time of non-water suppression is acquired (step 601). Here, a is an integer equal to or less than A representing a measurement number in the kx axis direction, b is an integer equal to or less than B representing a measurement number in the ky axis direction, c is an integer equal to or less than C representing a measurement number in the kz axis direction, In the case of two dimensions, one of A, B, and C is 1, and the other two are integers of 2 or more. In the case of 3 dimensions, all of A, B, and C are integers of 2 or more.

  Next, from the time-varying characteristics of the magnetic resonance signal Sw1 (kx (a), ky (b), kz (c)) (t (j)) in the k space obtained in step 601, the static magnetic field inhomogeneous time The change is calculated (step 602). In addition, as will be described later, it is assumed that this time variation of the static magnetic field inhomogeneity is caused by eddy current. Here, j is an integer equal to or less than J representing the data number in the time axis (t-axis) direction.

  Further, a magnetic field adjustment amount necessary for canceling the “magnetic field non-uniform time change” calculated in step 602 is calculated (step 603). Then, a magnetic resonance signal at the time of water suppression is measured while performing magnetic field adjustment based on the “magnetic field adjustment amount” calculated in step 603 (step 604).

  Finally, Fourier transform is performed on the magnetic resonance signal at the time of water suppression obtained in step 604, and a magnetic resonance spectrum image is calculated (step 605).

  Next, specific processing contents will be described using a data example obtained by “non-water suppression measurement” in step 601 in FIG. 5. FIG. 6 shows an example of measurement data in a two-dimensional real space, not a measurement example in a three-dimensional real space, in order to make the explanation of “non-water suppression measurement” in step 601 easier to understand. When measuring a two-dimensional real space, the application intensity of the phase encoding gradient magnetic field Gp1 may be fixed to zero (no change) in the MRSI sequence of FIG. Further, in the reference measurement (non-water suppression measurement) shown in FIG. 6, the other measurement conditions except for the presence or absence of water suppression are the same as the main measurement (water suppression measurement).

  Each image of (a), (b), (c), (d), (e) and (f) in FIG. 6 represents absolute value data in a two-dimensional k-space at different measurement times. The axis is the ky axis, the horizontal axis is the kx axis, and the luminance of each pixel luminance represents the signal intensity.

  FIG. 6G shows time-series data obtained by integrating absolute value data for all pixels, with the vertical axis representing signal intensity and the horizontal axis representing time. As shown in (g) of FIG. 6, the absolute value intensity of the measured time series signal is attenuated with the passage of time, and the attenuation rate is represented by a relaxation constant. FIG. 6A shows absolute value data in a two-dimensional k-space at the time of spin echo when static magnetic field inhomogeneity is canceled, and zero phase encoding points (of the phase encoding gradient magnetic fields Gp1 and Gp2 in FIG. 3) are shown. A pixel (star) having the maximum signal is located at the center of the image meaning that the applied intensity is zero.

  Comparing the images (a), (b), (c), (d), (e), and (f) in FIG. 6, the pixel (star) having the maximum signal is found as time passes. It can be confirmed that the image moves upward (in FIGS. 6 (a), (b), (c), (d), (e) and (f), it is easy to confirm the movement. In addition, “a quadrangle having a size of 8 pixels × 8 pixels” is superimposed at the same position in each image).

  These time-dependent changes in the images represent the accumulation of static magnetic field inhomogeneities as time passes. If the static magnetic field inhomogeneities are ideally zero, Except for the pixel having the maximum signal does not move from the center position of the image, and the temporal change as shown does not occur. For example, comparing the positions of the pixels (stars) having the maximum signals in FIGS. 6A and 6B, it can be seen that there is a shift of about one pixel in the ky axis direction.

  This is because the magnetic field inhomogeneity corresponding to one step of the phase encoding magnetic field Gp3 in the ky-axis direction is obtained in the time interval Tab from the measurement time Ta in FIG. 6A to the measurement time Tb in FIG. This means that the “component” has occurred, and that changes in these images over time reflect the static magnetic field inhomogeneity.

  Further, comparing the positions of the pixels (stars) having the maximum signals of (a) and (d) in FIG. 6, they are shifted by about 2 pixels in the ky axis direction. In addition, a “magnetic field inhomogeneous component corresponding to one step of the phase encoding magnetic field Gp3 in the ky axis direction” was generated in the time interval Tbd from the measurement time Tb to the measurement time Td in FIG. I understand.

  On the other hand, since the time interval Tbd is a period equivalent to twice the time interval Tab, it can be seen that the “magnetic field inhomogeneous component in the ky axis direction” is attenuated in terms of time. Therefore, the temporal change of these images also reflects the time change of the static magnetic field inhomogeneity, and it is assumed that the time change of the static magnetic field inhomogeneity is caused by the eddy current.

  As described above, the magnetic field inhomogeneity caused by the eddy current can be quantitatively estimated by observing the time change of the absolute value data in the two-dimensional k-space. It is possible to calculate the magnetic field adjustment amount necessary for canceling.

  A specific calculation example of “a magnetic field adjustment amount necessary for canceling the magnetic field inhomogeneity caused by the eddy current in the Y-axis direction” in the above measurement example will be shown below.

  First, the gradient magnetic field application amount Gp3s (time integral value) of “one step of the phase encode magnetic field Gp3 in the ky axis direction” is determined by the following equation (1) from the imaging conditions.

Gp3s = 2π / (γ · Fy) ――― (1)
Here, γ represents the magnetic rotation ratio of the hydrogen nucleus, and Fy represents the visual field in the Y-axis direction. As described above, “the magnetic field inhomogeneous component corresponding to one step of the phase encoding magnetic field in the ky axis direction” is generated at the Tab time interval, so that the Y axis direction having the gradient magnetic field strength (Gp3s / Tab) is generated. By applying the gradient magnetic field only for the period of Tab, it is possible to cancel the magnetic field inhomogeneity.

  Similarly, as described above, since the “magnetic field inhomogeneous component corresponding to one step of the phase encoding magnetic field Gp3 in the ky axis direction” is generated at the time interval Tbd, the gradient magnetic field strength (Gp3s / Tbd) is By applying the gradient magnetic field in the Y-axis direction for a period of Tbd, it is possible to cancel the magnetic field inhomogeneity. Here, since Tbd = Tab × 2, the gradient magnetic field strength applied in the Y-axis direction at the time interval Tbd may be half the gradient magnetic field strength applied at the time interval Tab.

  In the above calculation example, the calculation example in which the time interval Tab is used as the basic unit as the integration time when calculating the magnetic field inhomogeneity has been described, but this basic unit is narrowed to the sampling interval of the magnetic resonance signal. It is possible to perform more detailed magnetic field adjustment.

  When calculating the magnetic field adjustment amount, as described above, the magnetic field inhomogeneity amount estimated from the measurement conditions and the signal change amount and its time change, and the current change of each gradient coil and shim coil set at the time of design. Although it may be obtained by calculation from the characteristic of “quantity” versus “time-dependent change in generated magnetic field distribution”, it may be determined with reference to actually measured data according to the following procedure.

  FIG. 7 shows a magnetic field correction procedure (setting of the magnetic field adjustment amount) with reference to the actually measured data. As shown in step 801 of FIG. 7, first, prior to the reference image measurement, measurement of the “current change amount of gradient magnetic field coil and shim coil” versus “time-dependent change of generated magnetic field distribution” characteristic is performed for each gradient magnetic field coil and This is done for each shim coil. The spatial resolution and temporal resolution when measuring the spatial distribution and changes over time are preferably matched with the measurement conditions of the reference image measurement, but can be adjusted by post-processing (data interpolation or extrapolation). It is. In general, the temporal change rate of the magnetic field distribution generated by the high-order shim coil is often slower than the temporal change rate of the magnetic field distribution generated by the low-order shim coil or the gradient magnetic field coil. You may limit only to a next shim coil or a gradient magnetic field coil.

  Next, as shown in step 802 of FIG. 7, the measurement result of the “current change amount of each gradient coil and shim coil” versus “time-dependent change of generated magnetic field distribution” measured in step 801 is stored. Then, after performing the above-described reference image measurement, a time change of non-uniform magnetic field is calculated (step 803). Further, as shown in step 804 of FIG. 7, the time variation of the magnetic field inhomogeneity and the stored “current change amount of each gradient magnetic field coil and each shim coil” vs. “temporal change of generated magnetic field distribution” characteristics are measured. From the result, “the amount of current change to be passed through each gradient coil and shim coil” is calculated.

  If necessary, the calculated “current change amount to be passed through each gradient coil and shim coil” is stored (step 805), before water suppression measurement (main measurement) and before water suppression measurement (main measurement). During the process, the setting (control) of “the amount of change in current flowing through each gradient magnetic field coil and each shim coil” is performed (step 806).

  In step 804 in FIG. 7, when calculating “the amount of current change to be made to flow through each gradient coil and shim coil”, the distribution of the measured magnetic field inhomogeneity at each time and “each gradient magnetic field” are calculated. Each current value may be determined so that the difference in the combined magnetic field distribution obtained by synthesizing the magnetic field that can be generated by the coil and each shim coil (considering the change speed) is minimized.

The calculation of the time variation of the static magnetic field inhomogeneity and the calculation of the magnetic field adjustment amount shown in FIGS. 5 and 7 are performed by the computer 9.
As mentioned above, although the embodiment of the MRI apparatus to which the present invention is applied has been described, the present invention is not limited to the above-described embodiment, and various modifications and applications are possible. For example, in the above example, an example in which absolute value data in two-dimensional k-space is used as a reference image has been described. However, as described below, phase value data in two-dimensional real space is used as a reference image. It may be used.

  Each image of (a), (b), (c), (d), (e), and (f) in FIG. 8 represents phase value data in a two-dimensional real space at different measurement times. The axis is the Y axis, the horizontal axis is the X axis, and the luminance of each pixel luminance represents the phase value.

  Further, (g) in FIG. 8 represents time-series data obtained by integrating the absolute value data for all pixels, as in (g) in FIG. 6, where the vertical axis represents signal intensity and the horizontal axis represents time elapsed. Represents.

  FIG. 8A shows the phase value data in the two-dimensional real space at the spin echo time, and has a substantially uniform phase distribution. Comparing the images (a), (b), (c), (d), (e), and (f) in FIG. 8, the diagonal line shape from the upper right part toward the lower left part as time passes. It can be seen that the light and dark intervals of the striped pattern become narrower. In FIGS. 8A to 8F, in order to make it easier to confirm the state of phase distribution fluctuation, a “square having a size of 9 pixels × 9 pixels” is superimposed on the same position in each image. ing. Further, since the bright and dark stripes in FIGS. 8A to 8F are caused by aliasing due to phase discontinuity, the MRI phase distribution image described in “Background Art” is used. The discontinuity can be eliminated by using an aliasing removal technique generally used in the used shimming method or the like.

  8A to 8F, the static magnetic field inhomogeneity accumulates over time as in the time-dependent change of the absolute value data in the two-dimensional k-space shown in FIG. It represents the time variation of the state and static magnetic field inhomogeneity, and these images can also be used as the reference image described in the above example.

  In the above example, the case where the absolute value data in the two-dimensional k space and the phase value data in the two-dimensional real space are used as the reference image has been described. However, the absolute value data in the three-dimensional k space or 3 By using the phase value data in the three-dimensional space, it is possible to reduce the influence of eddy currents in the three-dimensional space.

  In the above example, the measurement conditions for the reference measurement (non-water suppression measurement) are the same as the main measurement (water suppression measurement) (excluding the presence or absence of water suppression). However, the measurement conditions are as follows: By changing a part of the measurement time, it is possible to shorten the measurement time required for the reference measurement.

  For example, in the MRSI pulse sequence shown in FIG. 3, when the number of steps of change in applied intensity of the phase encoding gradient magnetic fields Gp1, Gp2 and Gp3 is reduced by half, the measurement time is 1/8 (= 1/2 × 1 / 2 × 1/2). For example, in the example shown in FIG. 6, only data in “a quadrangle having a size of 8 × 8 pixels” superimposed at the same position in each image is measured.

  In the example shown in FIG. 7, when the number of applied intensity change steps is reduced, the spatial resolution in real space deteriorates. However, using the technique called zero filling described below, The above spatial resolution can be improved.

  A specific processing procedure of zero filling will be briefly described with reference to FIG. For example, when the number of phase encodings at the time of the main measurement (water suppression measurement) is 16 × 16 (two-dimensional space), the reference encoding (non-water suppression measurement) is performed with the phase encoding number of 8 without changing the phase encoding step. By reducing to x8 (two-dimensional space), the time required for reference image measurement can be reduced to 1/4 (= 1/2 x 1/2). However, the pixel size of the image obtained through the reconstruction process is four times (= 2 × 2).

  Here, the measured 8 × 8 data (8 × 8 time series signal) shown in (a) of FIG. 9 is arranged in the center of the 16 × 16 matrix shown in (b) of FIG. After the zero data is packed in the data missing portion, the inverse Fourier transform is performed in the direction of each phase encode axis, thereby making it possible to make the pixel size the same as when the number of phase encodes is 16 × 16. This is the zero filling process.

  Note that when such a zero filling process is performed, the apparent pixel size of the image obtained through the reconstruction process is reduced, but a high-frequency component that is originally necessary for increasing the spatial resolution is missing. Compared to a reconstructed image that is actually measured in full size (16 × 16), only an image with blurred spatial information can be obtained. That is, only the apparent spatial resolution is improved, but the actual spatial resolution is not improved.

  Regarding zero filling processing, when placing measurement data in the reconstruction matrix, the true zero encoding point (point having the maximum signal intensity on the spatial distribution) of the measurement data is detected and centered, and measurement is performed. Processing such as correction may be performed so that the data at the measurement center has the maximum signal intensity by multiplying the data by a Hamming function or the like on the spatial distribution.

  As another method for reducing the time required for reference measurement, there is a method for shortening the measurement repetition time. For example, when the repetition time during the reference measurement is reduced to half that during the main measurement, the measurement time can be reduced to ½. However, if the repetition time is simply shortened, the initial state of nuclear magnetization may change due to the influence of the longitudinal relaxation time, and correct phase information may not be obtained. In such a case, the flip angle of the excitation high-frequency magnetic field used in the non-water suppression measurement is set smaller than the flip angle of the water suppression measurement. By doing this, even if the repetition time is shortened, the influence of the longitudinal relaxation time can be reduced.

  Further, since the echo time (TE in FIG. 3) or the signal detection time (measurement time of the signal Sig1 in FIG. 3) is long, the repetition time may not be simply shortened. In such a case, the echo time (TE) used in the non-water suppression measurement is set shorter than the echo time (TE) of the water suppression measurement. That is, the detection start time of the magnetic resonance signal generated from the subject at the time of non-water suppression measurement is advanced from that at the time of water suppression measurement. Thus, the repetition time can be shortened.

  Further, the signal detection time used in the non-water suppression measurement can be set shorter than the signal detection time of the water suppression measurement, and the repetition time can be shortened.

  In the above example, the MRSI measurement is performed using the region selective spin echo type pulse sequence shown in FIG. 3 as an example, but the pulse sequence is not limited to this. The same effect can be obtained for all pulse sequences capable of MRSI measurement. For example, even when a measurement sequence called 3D-CSI or 4D-CSI that does not perform region selection, a high-speed MRSI sequence using an oscillating gradient magnetic field called EPSI, or the like is used, the same effect can be obtained.

  The eddy current correction method according to the embodiment of the present invention may be performed continuously after the static magnetic field uniformity correction using the modified spin echo method described in the above-mentioned “prior art”. Since the eddy current correction method according to the invention also has a correction effect on the static magnetic field uniformity without change over time, the present invention does not perform the static magnetic field uniformity correction using the modified spin echo method described in the above-mentioned “prior art”. Only the eddy current correction method according to the invention may be carried out.

  As described above, according to the present invention, in the non-water suppression measurement, the time variation of the static magnetic field inhomogeneity due to the eddy current is calculated, and the magnetic field adjustment amount necessary for canceling the calculated time variation of the static magnetic field inhomogeneity is calculated. calculate. Then, the main measurement (water suppression measurement) is performed while adjusting the magnetic field according to the calculated magnetic field adjustment amount.

  Therefore, since the main measurement is performed in a state where the static magnetic field inhomogeneity is reduced, a highly accurate image can be obtained as compared with the conventional technique in which the main measurement is performed in a state where the static magnetic field inhomogeneity is not sufficiently reduced. Can do.

  In addition, in the prior art, when static magnetic field inhomogeneity correction by eddy current is performed by data post-processing, it is necessary to perform reference measurement under the same measurement conditions as the main measurement in order to perform sufficient correction, and the total measurement time is long. Become. On the other hand, in the present invention, even if the reference measurement is shorter than the main measurement, high-accuracy data can be obtained without performing data post-processing, and the total measurement time can be shortened. .

  Next, the effects of the present invention described above compared with the prior art will be described. In order to obtain the result of eddy current correction, a pulse sequence shown in FIGS. 6 and 3 was executed using a magnetic resonance imaging apparatus of a static magnetic field strength of 1.5 Tesla of the type shown in FIG. MRSI measurement was performed. At this time, the target nuclide was a proton, and the measurement target was an N-acetylalanine (NAA) aqueous solution phantom.

  First, MRSI data measured when only the static magnetic field homogeneity correction using the modified spin echo method described in the “Background Art” section is performed (when the above-described eddy current correction is not performed) are shown in FIG. Shown in Here, FIG. 10 shows an integrated spectrum in the selective excitation voxel, and measurement took 8 minutes and 32 seconds (encoding number = 16 × 16, repetition time = 2 seconds).

  Next, FIG. 11 shows a correction result obtained by applying the “eddy current correction method using data post-processing” described in the “background art” column to the data shown in FIG. When the peak width of the spectrum in FIG. 11 (half width of the third peak from the right = 4.9 Hz) is compared with the peak width of the spectrum in FIG. 10 (8.8 Hz), “eddy current using data post-processing” It can be seen that the peak width is narrowed by the effect of the “correction method” (distortion is corrected).

  Since measurement of reference data for correction requires 8 minutes and 32 seconds (the number of encodings = 16 × 16, repetition time = 2 seconds), the total measurement time is 17 minutes and 4 seconds.

  FIG. 12 shows MRSI data measured after the eddy current correction of the embodiment of the present invention. When the peak width of the spectrum shown in FIG. 12 (3.9 Hz) is compared with the peak width of the spectrum shown in FIG. 10 (8.8 Hz), the peak width is narrow due to the effect of the eddy current correction of the present invention. You can see that

  Further, when the peak width of the spectrum shown in FIG. 12 is compared with the peak width of the spectrum shown in FIG. 11, it can be seen that the peak widths are almost equivalent. That is, according to the present invention, a correction effect equivalent to that obtained by performing data post-processing can be obtained without performing data post-processing.

  Here, in the example shown in FIG. 10, the total measurement time is 17 minutes 4 seconds. However, in the present invention, the measurement of the reference data for correction requires 2 minutes 8 seconds (encoding). (Number = 8 × 8, repetition time = 2 seconds), and the total measurement time is 10 minutes and 40 seconds.

  Therefore, it is possible to acquire a high-accuracy image in a short time without performing data post-processing according to the prior art.

1 is an external view of a magnetic resonance imaging apparatus to which the present invention is applied. 1 is a schematic block diagram of a magnetic resonance imaging apparatus that is an embodiment of the present invention. It is a figure which shows an example of a MRSI measurement sequence. It is a figure which shows an example of the pre-pulse sequence for water signal suppression in one Embodiment of this invention. It is a flowchart which shows the outline | summary of the measurement procedure of embodiment of this invention. It is a figure which shows the example of data (absolute value data of two-dimensional k space) obtained by non-water suppression measurement. It is a flowchart which shows the correction | amendment procedure in the case of referring the actual measurement data in one Embodiment of this invention. It is a figure which shows the example of data (phase value data of a two-dimensional real space) obtained by non-water suppression measurement. It is a figure for demonstrating a zero filling process. It is a figure which shows the MRSI data example measured when eddy current correction | amendment was not performed. It is a figure which shows the correction result which performed "the eddy current correction using data post-processing" to the data shown in FIG. It is a figure which shows the MRSI data measured after performing the eddy current correction by one Embodiment of this invention.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Subject, 2 ... Static magnetic field coil, 3 ... Gradient magnetic field coil, 4 ... Shim coil, 5 ... High frequency coil for transmission, 6 ... High frequency coil for reception, 7 ... Transmitter, 8 ... Receiver, 9 ... Computer, 10 ... Display, 11 ... Storage device, 12 ... Gradient magnetic field power supply, 13 ... Shim power supply, 14 ... Sequence control device, 15 ... Input device

Claims (10)

Static magnetic field generation means, gradient magnetic field generation means, high frequency magnetic field generation means, static magnetic field adjustment means for adjusting the spatial uniformity of the static magnetic field, and magnetic resonance signal detection means for detecting a magnetic resonance signal generated from the subject A magnetic field control unit that controls a gradient magnetic field generation unit, a high-frequency magnetic field generation unit, a static magnetic field adjustment unit, and a magnetic resonance signal detection unit to suppress a nuclear magnetic resonance signal from water and measure a nuclear magnetic resonance spectrum image. In a resonance imaging apparatus,
The control means calculates and calculates the amount of change over time in the measurement period for the spatial distribution of the nuclear magnetic resonance signal (non-water suppression magnetic resonance signal) measured without suppressing the nuclear magnetic resonance signal from water. Calculating a static magnetic field adjustment amount for canceling a static magnetic field inhomogeneous time change caused by an eddy current using the temporal change amount, and controlling the static magnetic field adjustment means using the calculated static magnetic field adjustment amount A magnetic resonance imaging apparatus. 2. The magnetic resonance imaging apparatus according to claim 1, wherein the control means uses a distribution of absolute value data in the k-space of the non-water suppression magnetic resonance signal as a spatial distribution of the non-water suppression magnetic resonance signal. Magnetic resonance imaging device. 3. The magnetic resonance imaging apparatus according to claim 2, wherein the control unit calculates the temporal change amount based on a change in a pixel position where the absolute value data in the k space is maximum. . 2. The magnetic resonance imaging apparatus according to claim 1, wherein the control means uses a distribution of phase value data in a real space of the non-water-suppressed magnetic resonance signal as a spatial distribution of the non-water-suppressed magnetic resonance signal. Magnetic resonance imaging device. Static magnetic field generation means, gradient magnetic field generation means, high frequency magnetic field generation means, static magnetic field adjustment means for adjusting the spatial uniformity of the static magnetic field, and magnetic resonance signal detection means for detecting a magnetic resonance signal generated from the subject A magnetic field control unit that controls a gradient magnetic field generation unit, a high-frequency magnetic field generation unit, a static magnetic field adjustment unit, and a magnetic resonance signal detection unit to suppress a nuclear magnetic resonance signal from water and measure a nuclear magnetic resonance spectrum image. In a resonance imaging apparatus,
The control means calculates information based on a nuclear magnetic resonance signal (non-water suppression magnetic resonance signal) measured without suppressing a nuclear magnetic resonance signal from water at a plurality of time points in time series, Using the information, the static magnetic field adjustment amount for a plurality of periods for canceling the time variation of the static magnetic field inhomogeneity caused by the eddy current is calculated, and the static magnetic field adjustment amount for the calculated plural periods is used to calculate the static magnetic field adjustment amount. A magnetic resonance imaging apparatus for controlling a magnetic field adjusting means . 6. The magnetic resonance imaging apparatus according to claim 5, wherein the plurality of periods are periods between the time points when the information is calculated . 7. The magnetic resonance imaging apparatus according to claim 1, wherein the control unit changes current of the gradient magnetic field generation unit and the static magnetic field adjustment unit prior to calculating the time variation of the static magnetic field inhomogeneity. The time variation characteristics of the magnetic field and the generated magnetic field distribution are calculated, and the time variation of the static magnetic field inhomogeneity is calculated based on the calculated time variation characteristics of the current variation and generated magnetic field distribution and the time variation of the static magnetic field inhomogeneity. A magnetic resonance imaging apparatus characterized by calculating a static magnetic field adjustment amount for canceling . The magnetic resonance imaging apparatus according to claim 1, wherein the measurement performed without suppressing the nuclear magnetic resonance signal from water is performed by suppressing the nuclear magnetic resonance signal from water. On the other hand, by reducing the number of phase encodings and adding zero data, the pixel size is set to the same pixel size as the measurement performed by suppressing the nuclear magnetic resonance signal from the water. Equipment . The magnetic resonance imaging apparatus according to any one of claims 1 to 8,
The static magnetic field adjusting means includes a shim coil,
The magnetic resonance imaging apparatus characterized in that the static magnetic field adjustment amount is a current change amount to be passed through the gradient magnetic field generating means and / or the shim coil . The magnetic resonance imaging apparatus according to any one of claims 1 to 9,
A magnetic resonance imaging apparatus using a spin echo type pulse sequence as a pulse sequence for measuring the nuclear magnetic resonance spectrum image by suppressing the nuclear magnetic resonance signal from the water.

Images