JPH1033495A - Nuclear magnetic resonance shooting method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce an echo time (TE) in split EPI. SOLUTION: In the split EPI, continuous reversal read-out inclined magnetic field pulse forms a trapezoid wave of identical height where time length of at least first pulse is shorter than that of others, substantially approx. 1/2-1/3. There at least signals of zero phase encode are partially measured by the first echo detected respectively by two measure sequences and all the data of zero phase encode are gained through several first echoes compensating one another.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a nuclear magnetic field for measuring a nuclear magnetic resonance (hereinafter, referred to as NMR) signal from hydrogen, phosphorus or the like in an object to visualize a nuclear density distribution, a relaxation time distribution and the like. The present invention relates to a resonance imaging (MRI) method, and particularly to an improvement of an echo planar method (EPI).

[0002]

2. Description of the Related Art In recent years, an echo planar sequence (EPI) as shown in FIG. 6 has been frequently used as a high-speed MRI sequence. In the EPI sequence, a high-frequency pulse 201 is applied to a subject including magnetization to be detected, and at the same time, a gradient magnetic field pulse 202 for selecting a slice is applied. Thereby, the slice to be imaged is selected. Next, a pulse 203 for giving a phase encoding offset and a pre-pulse 205 for giving a read gradient magnetic field offset are applied.
After that, the readout gradient magnetic field pulse that continuously reverses
Apply 206. The gradient magnetic field pulse 206 is a trapezoidal pulse. A phase encoding gradient magnetic field pulse 204 (solid line) is discretely applied in synchronization with the gradient magnetic field pulse 206. Within each period of the inverted readout gradient magnetic field 206, the echo signal 207 of each phase encoding is generated in a time-series sequence, and is sampled for each time range 208 to obtain time-series data. Each echo 207 is usually obtained as a time-series signal composed of 128, 256, 512, 1024, etc. sampling data. Typically, each sampling time range 208 requires about 1 ms. The interval between adjacent time ranges 208 is about 0.5 ms to 1 ms. That is, the time length of the inversion gradient magnetic field pulse 206 needs to be about 1.5 to 2 ms in order to obtain predetermined time-series data, and the pre-pulse 205 is a pulse having a half time length of the inversion gradient magnetic field pulse 206. Is given.

As described above, in MRI, the gradient magnetic field pulse G
Different phase encodings are given by e, and echo signals obtained by each phase encoding are detected. As the number of the phase encodes 204 (that is, the number of the echo signals 207), values such as 128, 256, and 512 are usually selected for one image.
Through such a series of operations 209, all echoes necessary for image reconstruction are collected, and each obtained echo signal group (for example, time series data of 265 data × 256 echoes) is subjected to two-dimensional Fourier transform (FFT). To create one MR image. In this way, all echoes are collected by one high-frequency pulse irradiation E
The PI sequence is specifically called one-shot EPI.

The cross section to be photographed is a gradient magnetic field in the slice direction.
If it is determined in 202 and the same section is continuously photographed, after waiting for time 210 to wait for the recovery of magnetization, operation 209
repeat. This latency 210 typically ranges from 1 second to 2
In seconds, the time required for operation 209 is on the order of 100 ms.

[0005] In this one-shot EPI, since all echoes are collected by one high-frequency pulse irradiation, there is an advantage that the photographing time is extremely short, but the intensity of the echo signal decreases as time elapses from the high-frequency pulse irradiation. However, there is a problem that an image with good SN cannot be obtained. A split EPI has been proposed to solve this problem.

In the divided EPI, only a part of the phase encoded data is obtained in the operation 209 shown in FIG. 6, and the operation 209 (indicated by a dotted line in the figure) is repeated while changing the pulse 203 for giving the phase encoding offset. The remaining echo signal 207 is obtained. The figure shows an example of two divisions.
In the first repetition, a pre-pulse 203 and a phase encode pulse 204 whose polarity is inverted are applied. The split EPI is
Since the operation 209 is repeated, the data acquisition time for acquiring one image becomes longer, but the number of echoes 207 acquired in the operation 209 decreases, and the operation 209 itself becomes shorter. As a result, signals can be acquired before the attenuation of the echo 207 becomes significant, and high image quality can be achieved.

[0007] The MRI signal acquisition procedure is generally represented by those skilled in the art using a k-space trajectory as shown in FIG. In the k-space, the horizontal axis indicates the readout gradient magnetic field application amount,
The vertical axis indicates the amount of phase encoding. The coordinates of the k space are represented by (kx, ky), and the center of the space is the origin (0, 0). FIGS. 7A and 7B show one-shot E, respectively.
This figure shows the locus of k-space in the case of PI and in the case of divided EPI. In the figure, a black circle is a data collection start position. As is clear from the figure, data obtained by sampling one echo is arranged on the trajectory as complex data (detection signal itself) along the horizontal axis of k-space in the order of data acquisition. A data string for one echo corresponds to exactly one horizontal row of the trajectory. Each time a phase encode pulse is applied, the locus shifts in the vertical axis direction. One shot EPI
Then, the space is scanned in a one-stroke form in the k space. In the case of the two-split EPI, the upper half of the k space is scanned with the first echo train (data obtained by one measurement), and the lower half of the k space is scanned with the second echo train. After performing known inversion processing and phase correction processing on the data on the scanning line,
By performing the two-dimensional FFT, an image in a real space is obtained.

[0008]

In an MR image, the time from RF irradiation until acquisition of data at the origin of k-space is called an effective TE (echo time).
Is an important parameter that determines the contrast of an image. Generally, shortening the effective TE has many advantages. For example, when the effective TE is shortened to about 2 to 3 ms, in a fluid such as blood, signal loss due to turbulence is reduced. Also, signal attenuation due to longitudinal relaxation is reduced.

In the split-type EPI, the effective TE can be shortened by assigning a phase encode amount of zero to the first echo as the phase encode application order (see FIG. 7). However, even in this case, a method of shortening the effective TE to about 1 ms has not been proposed.

On the other hand, there is an imaging sequence called a spiral scan method in which k-space is sampled on a spiral in addition to EPI. In this method, data is acquired starting from the origin of k-space, so that the above-mentioned effective TE is shortened. It is possible to However, since the spiral scan method has a special sampling method, it is necessary to always interpolate the data, and the EPI such as deterioration of image quality due to uneven magnetic field.
The various merits of cannot always be exhibited.

Accordingly, an object of the present invention is to further shorten the effective TE in the EPI sequence and to obtain an image having a contrast and image quality different from those of the related art.

[0012]

In order to achieve the above object, the MRI method of the present invention irradiates a high-frequency pulse to a subject containing a magnetization to be detected to generate transverse magnetization, and reads out the subject to which the transverse magnetization has been applied. After giving the gradient magnetic field offset amount, while applying the reading gradient magnetic field pulse which inverts continuously, the phase encoding gradient magnetic field pulse is applied in synchronism with this, and it is generated within each period of the reading gradient magnetic field pulse which inverts. In a nuclear magnetic resonance imaging method for detecting an echo signal as time-series data and reconstructing an image from these time-series data groups, the readout gradient magnetic field pulse is provided as a continuous inversion pulse having the same height, and
Is shorter than the pulse time length (t0) required to acquire time series data of one echo. In this case, the readout gradient magnetic field offset amount is also set to be smaller than the normal pre-pulse corresponding to the first inversion gradient magnetic field pulse having a short time length. The read gradient magnetic field offset may be zero.

Further, in the MRI method of the present invention, the readout gradient magnetic field pulse Gr is given as a series of inverted trapezoidal waves having the same height, and as shown in FIG. 2, at least the time length of the first trapezoidal wave pulse ( t) is shorter than the time length (t0) of the remaining trapezoidal wave, and specifically, is set to 1/2 to 2/3 of the time length of the remaining trapezoidal wave. With such a range, the effect of shortening the TE time becomes remarkable.
When the readout gradient magnetic field offset amount is zero, at least the first trapezoidal wave Hals can be set to １ ／ of the time length of the remaining trapezoidal wave pulses. In FIG. 2, the gradient magnetic fields in the phase encoding direction and the slice direction are omitted for the sake of simplicity.

After the irradiation of the high-frequency pulse, when a readout gradient magnetic field offset is given, the time integration value (pulse height × time) of the subsequently inverted first readout gradient magnetic field applied becomes equal to the offset amount. One echo is generated (FIG. 2). The first echo is an echo signal having zero phase encode, is sampled for a predetermined time, and is arranged as time-series data in the center column of the k-space. Here, as shown in FIG. 3B, the data sequence obtained by the normal echo signal occupies the entire width of one column of the coordinates in the k space, and the read gradient magnetic field offset amount and the time integration value of the next read gradient magnetic field are different. The data at the time point T when they become equal becomes the data at the center of the column. In contrast, the EPI of the present invention
Since the time length of the first readout gradient magnetic field is set to be short, the interval from the start of sampling to the time point T becomes short as shown in FIG. Such data is arranged with the coordinates closer to the center rather than the end of the k space as the start position, as shown in FIG. This means that the effective TE, which is the time from the irradiation of the high-frequency pulse to the acquisition of the data of the origin of the k space, has been reduced.

The readout gradient magnetic field offset amount can be set to zero. In this case, half the data of the echo train is acquired, and the effective TE can be minimized and can be reduced to about 1 ms.

According to the MRI method of the present invention, data with zero phase encoding can be measured only partially in the first echo signal. Therefore, this data is partially measured by the first echoes of different measurements, and the entire zero-phase encoded echo signal is acquired complementarily by the plurality of first echoes.
In this case, data near the origin including the origin of the k space may partially overlap.

The MRI method of the present invention can be applied to a one-shot EPI for measuring all echoes filling the k-space by one high-frequency pulse irradiation. This is applied to the divided EPI in which the measurement for acquiring the data is repeated a plurality of times. One shot EPI
In the case of (1), the effect of the present invention becomes remarkable by using the half Fourier imaging together, and the first data is obtained in order to obtain the complete data of the signal whose phase encoding is zero.
The echo measurement may be repeated. Further, the same sequence may be repeated, and the echo signals other than the zero-phase encoding may be added.

In the present invention, not only the first readout gradient magnetic field pulse but also a plurality or all of the inverted readout gradient magnetic field pulses are set to a pulse time length necessary to acquire all the time series data of one echo. It can be shorter than (t0). In this case, the echo is generated not at the center of the pulse period but alternately before or after, and a part of the time-series data is acquired. Such time-series data can be so-called half-echo imaging in which an image is reconstructed with half data using the complex conjugate property of the echo signal, but by repeating two or more times,
A plurality of sets of echo signals may be measured for each phase encode, and the entire set of phase encode signals may be obtained from the plurality of sets of signals.

[0019]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, the MRI method of the present invention will be further described with reference to examples.

FIG. 4 shows an outline of a typical MRI apparatus to which the MRI method of the present invention is applied. The MRI apparatus includes a magnet 402 for generating a static magnetic field around a subject 401,
A gradient magnetic field coil 403 that generates a gradient magnetic field in a measurement space where the subject 401 is placed, an RF coil 404 that generates a high-frequency magnetic field in this space, and an RF probe 405 that detects an NMR signal generated by the subject 401 are provided. ing. Gradient magnetic field coil 4
03 is composed of gradient magnetic field coils in three directions of X, Y and Z,
A gradient magnetic field is generated in accordance with a signal from the gradient magnetic field power supply 409. The RF coil 404 generates a high-frequency magnetic field according to the signal of the RF signal unit 410. The signal of the RF probe 405 is detected by the signal detection unit 406, processed by the signal processing unit 407, and converted into an image signal by calculation. The image is displayed on the display unit 408. Gradient magnetic field power supply 409, RF transmitter
410 and the signal detection unit 406 are controlled by the control unit 411 according to a predetermined pulse sequence.

In the MRI method of the present invention, an echo planar sequence (EPI) is employed as the pulse sequence. In particular, a part of a signal necessary for reconstructing one image is measured by one high-frequency pulse irradiation. The divided EPI is used to obtain the entire signal necessary for reconstructing one image by repeating the measurement twice.

An embodiment in which the present invention is applied to a two-part EPI will be described below with reference to FIG. In the figure, items common to FIG. 6 are denoted by the same reference numerals.

This embodiment is particularly different from the divided EPI of FIG. 6 in that the area of the pre-pulse 105 of the readout gradient magnetic field Gr is smaller than that of the pre-pulse 205 of FIG. 1061 has an area that is slightly more than half the area of 1062, 1063,.

The rest is almost the same as the divided EPI shown in FIG. 6, but briefly, first, a high frequency pulse 20
At the same time as irradiating 1, a gradient magnetic field pulse 202 for selecting a slice is applied. Next, a pre-pulse 105 shorter than the normal EPI is applied, and then a read gradient magnetic field 106 that is continuously inverted is applied. Then, the echo signal 207 is measured in each cycle of the readout gradient magnetic field to be inverted. A phase encoding gradient magnetic field 20 for phase encoding an echo signal in synchronization with the reversal of the reversing read gradient magnetic field Gr.
4 is applied. As a result, the echo signal 207 is given a phase encoding that is incremented by one phase encoding with the zero phase encoding as an initial value.

In the two-part EPI of the present embodiment, the upper half and the lower half of the k-space are scanned by repeating the illustrated sequence twice. A gradient magnetic field having a reversed polarity is applied. However, as will be described later, in the EPI of the present invention, the entire signal of the zero-phase encoding is acquired by two first echoes measured by two repetitions. Not done.

In this embodiment, the readout gradient magnetic field
106 is a trapezoidal wave whose polarity is inverted at the same height, and whose time length is shorter in the first readout gradient magnetic field 1061 alone than in the other readout gradient magnetic fields 1062, 1063. As described above, the time length (pulse width) of the reversal gradient magnetic field in the normal EPI sequence is required to be about 1.5 to 2 ms in order to obtain a required number of sampling data. Are the same, a pulse having a time length of about 1/2 of that is used. In the present embodiment, a pulse whose time length is shorter than 1/2 (can be shortened when the height is higher) is used as the pre-pulse. By using such a pulse, a pulse having a time length shorter than the pulse time length generally required for the first inverted readout gradient magnetic field 1061 can be obtained. For example, when the pre-pulse has a time length of 1/6 of the second and subsequent inversion gradient magnetic field pulses, the time length of the first inversion gradient magnetic field pulse 1061 is 2/3 (= 1) of the second and subsequent inversion gradient magnetic field pulses. / 6 + ／).
It is also possible to omit the pre-pulse (the pre-pulse is zero). In this case, the time length of the first inversion gradient magnetic field pulse 1061 can be reduced to half. Note that the time length of the pulse can be realized by changing the time of the flat portion of the trapezoidal wave.

In the echo signal, a first echo 1071 is generated at a time 100 when the time integration of the pre-pulse 105 and the time integration of the first inversion gradient magnetic field pulse 1061 become equal. This first echo is a signal of zero phase encoding, and a high frequency pulse
The time from 201 to the first echo generation (the time of the slice gradient magnetic field pulse + the time of the pre-pulse + the first half time of the inversion gradient magnetic field pulse) is the effective TE. Therefore, in the present embodiment, the pre-pulse 105 and the first gradient magnetic field pulse 1061 are set to have a time length of, for example, 1/6 and ／ of the second and subsequent gradient magnetic field pulses, respectively.
From the rise of the first echo to the generation of the first echo can be reduced to 1/3 of the conventional case, and the effective TE can be shortened. When the pre-pulse is set to zero, the first reversal gradient magnetic field pulse 1061 has a half time length of the second and subsequent gradient magnetic field pulses, and the effective TE is determined only by the time of the slice gradient magnetic field pulse. Become.

The time integration of the first inversion gradient magnetic field pulse 1061 and the time integration of the second inversion gradient magnetic field pulse 1062 after the first echo generation 100 are equal to the time integration of the second inversion gradient magnetic field pulse 1062. , A second echo occurs. Hereinafter, similarly, the third and fourth echoes are generated when the remaining time integrated value becomes equal to the time integrated value of the subsequent inverted pulse. These second and subsequent echoes all occur at the center of the reversal gradient magnetic field pulse, and the measurement (sampling) of the echo signal is performed as in the conventional EPI.

The echo signal is sampled in a predetermined time zone of the flat portion of the trapezoidal wave. Since the first echo has a short generation time as described above, the sampling time is correspondingly short. Therefore, the data obtained by sampling the first echo does not scan the entire width of the k space, but scans 2/3 of the full width in the example described above. Therefore, the rest is obtained by the first echo of the second sequence. This will be described with reference to the k-space trajectory shown in FIG.

First, in the first measurement in which a phase encoding gradient magnetic field is applied as shown by a solid line in the sequence of FIG. 1, the first echo measured after the application of the first inversion gradient magnetic field pulse 1061 is zero phase encoding. Data
Scan the middle row of k-space. Here, in the first sequence, the first echo is shorter in data length than the second and subsequent echoes. Therefore, the entire width is not scanned, and the start position is set near the center. As described above, since the first echo starts near the origin of the k-space, it can be understood that the origin data can be obtained in a short time. The phase encode then moves upward by incrementing by one, and the second echo
Scan over the echo. Hereinafter, similarly, the upper half of the k space is scanned while shifting the vertical coordinate every time the phase encoding is incremented by one. Next, in the second measurement, the lower half of the k-space is scanned by applying a phase encoding gradient magnetic field as shown by a dotted line in FIG. Here, even in the second time, the first echo of the zero-phase encoding is shorter than the subsequent echoes, and is located near the center without scanning the entire width. Therefore, all data of the zero phase encoding is measured without loss by complementing with each first echo of these two measurements.

When the pre-pulse is set to zero,
As for the echo generated by the first gradient magnetic field pulse 1061, only half the data of the echo generated in the period of the subsequent gradient magnetic field pulse is obtained, so that the data of the full width is obtained by the first measurement and the second measurement. Will be done. When a short pre-pulse is used and the time length of the first gradient magnetic field pulse 1061 is set to be longer than の of the time length of the other gradient magnetic field pulses, some data centering on the origin of k-space is used. Although it will overlap (the figure is slightly shifted in the vertical direction for easy viewing,
However, these data are added and used, so that the SN of the signal can be improved.

The data on the k space obtained as described above is subjected to inversion processing, phase correction processing, and the like according to a conventional method, and then subjected to two-dimensional FFT to obtain an image of a slice selected slice. Can be reconfigured. Further, by repeating and adding these EPI sequences as necessary, an image with improved SN can be obtained.

The case where the present invention is applied to a two-split EPI sequence has been described above, but the present invention can also be applied to a multi-shot EPI such as a four-split EPI. FIG. 5B shows a k-space trajectory when the MR method of the present invention is applied to a quadrant EPI. Here, the first to fourth echo trains are measured by repeating the measurement sequence (FIG. 1) four times. In this case, two rows near the zero-phase encoding are complementarily scanned by two echo trains, and full width scanning is performed. In the illustrated example, one row of data is scanned by the first and third echo trains, and another row of data is scanned by the second and fourth echo trains. There are no restrictions,
It can be changed arbitrarily.

[0034]

As is clear from the above embodiments, according to the MRI method of the present invention, the time length of at least the first readout gradient magnetic field pulse in the EPI sequence is set as follows.
The time length of the remaining readout gradient magnetic field pulse (the pulse time length required to acquire time series data of one echo) is shorter, preferably 1/2 to 2/3, and the entire signal of zero phase encoding Can be shortened by the time from irradiation of the high-frequency pulse to the acquisition of the first echo, that is, the execution TE. As a result, the effective TE can be shortened to about 2 to 3 ms, and the influence of signal loss due to turbulence is small even in imaging of a fluid such as blood, and the signal attenuation due to longitudinal relaxation is small. Images can be obtained.

[Brief description of the drawings]

FIG. 1 is a time chart showing a two-part EPI sequence according to one embodiment of the present invention.

FIG. 2 is a diagram illustrating an MRI method of the present invention.

3A and 3B are diagrams for explaining time-series data acquired by the MRI method of the present invention, wherein FIG. 3A shows the k-space arrangement of a zero-phase encoded signal according to the method of the present invention, and FIG. Diagram for explaining k-space arrangement of signals

FIG. 4 is a block diagram of an MRI apparatus to which the present invention is applied;

5A and 5B are diagrams illustrating k-space arrangement of time-series data acquired according to the embodiment of the present invention, wherein FIG. 5A illustrates a case using the 2-split EPI method, and FIG. 5B illustrates a case using the 4-split EPI method.

FIG. 6 is a time chart showing a conventional EPI sequence.

FIG. 7 is a diagram showing data arrangement in k-space by the conventional EPI method.

[Explanation of symbols]

 106: Readout gradient magnetic field pulse 201: High frequency pulse 204: Phase encoding gradient magnetic field pulse 207: Echo signal

Claims (4)

[Claims] 1. A readout gradient that irradiates a high-frequency pulse to a subject including a magnetization to be detected to generate transverse magnetization, applies a readout gradient magnetic field offset to a subject to which transverse magnetization is applied, and then continuously reverses the readout gradient. While applying a magnetic field pulse,
In synchronization with this, a phase encoding gradient magnetic field pulse is applied, an echo signal generated in each cycle of the inverted read-out gradient magnetic field pulse is detected as time-series data, and a nucleus for reconstructing an image from the time-series data group is detected. In the magnetic resonance imaging method, the readout gradient magnetic field pulse is provided as a continuous inversion pulse having the same height, and at least the time length of the first pulse is used to acquire time-series data corresponding to one echo. A nuclear magnetic resonance imaging method characterized by being shorter than a required pulse time length. 2. A readout gradient in which a subject containing a magnetization to be detected is irradiated with a high-frequency pulse to generate transverse magnetization, a readout gradient magnetic field offset is applied to the subject to which transverse magnetization is applied, and then the readout gradient is continuously inverted. While applying a magnetic field pulse,
In synchronization with this, a phase encoding gradient magnetic field pulse is applied, an echo signal generated in each cycle of the inverted read-out gradient magnetic field pulse is detected as time-series data, and a nucleus for reconstructing an image from the time-series data group is detected. In the magnetic resonance imaging method, the read-out gradient magnetic field pulse is given as a continuous inversion pulse having the same height, and each pulse is divided into at least two or more types of groups with the time length.
Wherein the time length of the pulse of the groom including the inversion pulse is shorter than the time length of the pulses of the other groups. 3. The nuclear magnetic resonance imaging method according to claim 1,
The measurement sequence from the irradiation of the high-frequency pulse to the detection of the echo signal is repeated twice or more, and at least the signal of the zero-phase encode is partially measured by the first echo detected in each of the two measurement sequences. A nuclear magnetic resonance imaging method in which all encoded data are acquired complementarily with a plurality of first echoes. 4. The read gradient magnetic field pulse has a pulse time length required for acquiring time-series data corresponding to one echo, wherein the time length of the first pulse or a pulse of a group including the first pulse is set. The nuclear magnetic resonance imaging method according to claim 1 or 2, wherein the thickness is 1/2 to 1/3.