 |
Description  |
|
|
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention is directed to a method for operating a nuclear
magnetic resonance tomography apparatus, and in particular to a method for
rapidly generating a magnetic resonance image with improved diagnostic
utility.
2. Description of the Prior Art
Magnetic resonance imaging devices are known in the art which have a first
set of coils for generating a fundamental or static magnetic field in
which an examination subject is disposed, a plurality of further coils for
respectively generating gradient magnetic fields in which the examination
subject is also disposed, and a coil or antenna for irradiating the
examination subject with a sequence of radio-frequency (RF) pulses and for
acquiring the resulting nuclear magnetic resonance signals from the
examination subject. In such devices, after each RF pulse, a first
gradient pulse in a first direction is generated as a de-phasing gradient,
and another gradient pulse in at least one second direction is then
generated as a first phase-coding gradient. Following the first
phase-coding gradient, a first signal, occurring during a second gradient
pulse inverted relative to the first gradient pulse, is read-out as a
first read-out gradient.
So-called fast imaging sequences have recently been developed for
minimizing the time required to obtain the necessary signals to construct
a magnetic resonance image. A so-called FLASH sequence is described, for
example, in "Rapid Images and NMR-Movies", Haase et al, SMRM, Fourth
Annual Meeting, Book of Abstracts (1985), page 980. In this sequence,
excitation of nuclear magnetic resonance signals with RF pulses is
undertaken with flip angles substantially below 90.degree., with the
fastest possible repetition rate. The image contrast is defined by the
longitudinal relaxation time T.sub.1.
Another sequence, known as FISP, is described in "FISP-A New MRI Sequence,"
Oppelt et al, Electromedica 54, Vol. 1, 1986. In contrast to the FLASH
sequence, the phase memory of the spin system is exploited in the FISP
sequence by refocusing the coding gradient to obtain an image which is
determined by the ratio of the longitudinal relaxation time to the
transverse relaxation time (T.sub.1 /T.sub.2).
A FISP sequence of the type utilized in practice is shown in FIG. 2 herein.
To obtain nuclear magnetic resonance, the examination subject is exposed
to a sequence of RF pulses. A slice selection gradient G.sub.z1 is
generated simultaneously with each RF pulse, so that only the slice of the
examination subject defined by its z-coordinate is excited. After such
excitation has occurred, a gradient pulse G.sub.z2, having a negative
direction in comparison to the slice gradient G.sub.z1, is generated. The
dephasing of the spin system caused by the slice selection gradient
G.sub.z1 is thus cancelled.
A negative gradient pulse G.sub.y1 and a phase coding gradient G.sub.x1 are
generated simultaneously with the gradient pulse G.sub.z2. The spin system
is de-phased with the gradient pulse G.sub.y1, and the FID signal
following the RF pulse is thus destroyed. The phase coding gradient
G.sub.x1 is generated after each RF pulse, and impresses phase information
on the spin system. The negative gradient pulse G.sub.y1 is followed by a
positive gradient pulse G.sub.y2, which in turn cancels the de-phasing
caused by the gradient pulse G.sub.y1, and thus generates a spin echo and
thus an interpretable signal S.sub.1. After the decay of this signal, a
second phase coding gradient G.sub.x2 is generated, which is inverted
compared to the first phase coding gradient G.sub.x1. The de-phasing
caused by the first phase coding gradient G.sub.x1 is thus cancelled.
A modified FISP sequence known as CE FAST (shown in FIG. 3 herein) was
presented by Gingel et al the SMRW Conference in 1986 in Montreal, and is
described in "The Application Of Steadystate Free Precession (SFP) in
2D-FT MR Imaging," Gingel et al, SMRM, 5th Annual Meeting, Book of
Abstracts (1986), page 666. Again, RF pulses are generated in rapid
succession simultaneously with a slice selection gradient G.sub.z1. Each
slice selection gradient G.sub.z1 is preceding by a negative gradient
pulse G.sub.z3, which cancels the de-phasing caused by the slice selection
gradient G.sub.z1. Further, each RF pulse is preceded by a negative
gradient pulse G.sub.y4 and by a phase coding gradient G.sub.x1. Following
each RF pulse, a positive gradient pulse G.sub.y3, together with a
phase-coding gradient G.sub.x2 which is inverted in comparison to the
first phase-coding gradient G.sub.x1, are generated. A half echo signal,
which is shaped to form a full echo signal by the action of the gradient
pulses G.sub.y4 and G.sub.y3, thus arises, as shown in "Phase and
Intensity Anomalies in Fourier Transform NMR," Freeman et al, Journal of
Magnetic Resonance, Vol. 4, pages 366-383 (1971).
T.sub.1 /T.sub.2 -weighted images with respectively different information
content can be produced with the FISP sequence of FIG. 2, and T.sub.2
-weighted images having respectively different information content can be
produced with the CE FAST sequence of FIG. 3.
SUMMARY OF THE INVENTION
It is an object of the present invention to obtain the information content
of both the FISP sequence and the CE FAST sequence without lengthening the
measuring time in a nuclear magnetic resonance tomography apparatus.
The above object is achieved in accordance with the principles of the
present invention by using the known steps of generating a sequence of RF
pulses, generating a first gradient pulse G.sub.y1 in a first direction,
such as the y-direction, as a dephasing gradient and a first gradient
pulse G.sub.x1 in at least one second direction, such as the x-direction,
as a first phase-coding gradient, and generating a second gradient pulse
G.sub.y2, inverted relative to the first gradient pulse G.sub.y1 during
which a first signal S.sub.1 is read-out. Additionally, in accordance with
the principles of the present invention, a third gradient pulse G.sub.y3
in the first direction, such as the y-direction, is generated preceding
each RF pulse during which a second signal S.sub.2 is read-out, and a
fourth gradient pulse G.sub.y4, inverted relative to the third gradient
pulse G.sub.y3, is generated in the first direction, such as the
y-direction, following the third gradient pulse G.sub.y3 together with a
second gradient pulse G.sub.x2 in the second direction, for example the
x-direction, which is inverted relative to the first phase-coding gradient
G.sub.x1, and which serves as a second phase-coding gradient.
The first signal S.sub.1 obtained by this method is T.sub.1 /T.sub.2
-weighted, and the second signal S.sub.2 is T.sub.2 -weighted. Two images
having respectively different T.sub.2 -contrast, and thus improved
diagnostic utility, can thus be produced during a substantially
non-increased measuring time.
In a further embodiment of the method, resolution in a third dimension or
direction is achieved by generating a slice selecting gradient G.sub.x1 in
the third direction, such as the z-direction, simultaneously with each RF
pulse, and generating additional gradient pulses G.sub.z2 and G.sub.z3,
each having an opposite sign relative to the slice selecting gradient
G.sub.z1, respectively preceding and following each slice selecting
gradient G.sub.z1.
In a further embodiment, a three-dimensional data set can be registered by
generating respective phase coding gradients G.sub.z4 and G.sub.z5 in the
third direction, such as the z-direction, respectively preceding and
following each RF pulse, the phase-coding gradients G.sub.z4 and G.sub.z5
having opposite operational signs.
DESCRIPTION OF THE DRAWINGS
FIG. 1 is schematic block diagram of a nuclear magnetic resonance
tomography apparatus of the type for practicing the method disclosed
herein.
FIG. 2 is an exemplary depiction of a RISP sequence, as described above.
FIG. 3 is an exemplary embodiment of a CE FAST sequence, as described
above.
FIG. 4 is an excitation sequence for a method for operating a nuclear
magnetic resonance tomography apparatus of the type shown in FIG. 1
according to a first embodiment of the method disclosed herein.
FIG. 5 is an excitation sequence for operating a nuclear magnetic resonance
tomography apparatus of the type shown in FIG. 1 according to a further
embodiment of the method disclosed herein.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
The basic components of a nuclear magnetic resonance tomography apparatus
of the type in which the method disclosed herein may be used for producing
images of an examination subject are shown in FIG. 1. A static or
fundamental magnetic field is generated by coils 1, 2, 3 and 4 in which an
examination subject 5 is disposed, such as for examination for medical
diagnostic purposes. A plurality of gradient coils, in which the
examination subject 5 is also disposed, are provided for generating
orthogonal magnetic field gradients in the x-, y- and z-directions, as
indicated by the coordinate system 6. For clarity, only gradient coils 7
and 8 are shown in FIG. 1, which generate the x-gradient in combination
with a pair of identical gradient coils disposed opposite thereto. Sets of
coils (not shown) for generating the y-gradient are disposed parallel to
each other above and below the examination subject 5. Sets of coils (not
shown) for generating the z-gradient are disposed at the head and feet of
the examination subject 5 transversely relative to the longitudinal axis
of the subject 5.
The apparatus also includes a RF coil 9 which induces and acquires nuclear
magnetic resonance signals in and from the examination subject 5. The
coils 1-4 and 7-9, bounded by the dot-dash line 10, represent the actual
examination instrument in which the subject 5 is disposed.
The instrument 10 is operated by electrical components including a power
supply 11 for the coils 1-4 and a power supply 12 for the gradient coils 7
and 8 and the other gradient coils which are not shown in the drawing. The
coil 9 is connected either to a signal amplifier 14, for receiving nuclear
magnetic resonance signals from the examination subject 5, or to a RF
transmitter 15, for exposing the examination subject 5 to RF pulses, by a
switch 19 switchable between a transmission mode and a reception mode. The
amplifier 14 and the transmitter 15 are part of an RF unit, which is
connected to a process computer 17. The process computer 17 is also
connected to the gradient coils power supply 12 and controls the
generation of an image from the nuclear magnetic resonance signals. The
image constructed from these signals is supplied by the computer 17 to a
display 18 for viewing.
A first exemplary embodiment of a method in accordance with the principles
of the present invention for operating the apparatus shown in FIG. 1 to
obtain an image of the examination subject 5 is shown in FIG. 4. In this
embodiment, the examination subject 5 is exposed to a sequence of
radio-frequency pulses RF supplied simultaneously in the presence of a
slice selection gradient G.sub.z1. Each radio-frequency pulse RF is
preceded by negative gradient pulses G.sub.z3 and G.sub.y4, as well as by
a phase-coding gradient G.sub.x2. The phase-coding gradient G.sub.x2 is
varied in amplitude from radio-frequency pulse to radio-frequency pulse.
Each radio-frequency pulse RF is followed by negative gradient pulses
G.sub.z2 and G.sub.y1, as well as by a phase-coding gradient G.sub.x1,
which is inverted in comparison to the phase-coding gradient G.sub.x2. The
negative gradient pulse G.sub.y1 is followed by a positive gradient pulse
G.sub.y2, and the negative gradient pulse G.sub.y4 is preceded by a
positive gradient pulse G.sub.y3.
A state of equilibrium in the transverse magnetization of a lice of the
examination subject 5 is produced with the series of selected
radio-frequency pulses RF. As explained in the aforementioned article by
Freeman ("Phase and Intensity Anomalies in Fourier Transform NMR"),
focusing points of the transverse magnetization are formed immediately
preceding and following the radio-frequency pulses RF. These focusing
points can be imagined in a first approximation as an FID signal S.sub.+
and as a half echo signal S.sup.-. The FID signal S.sup.+ is de-phased by
the negative gradient pulse G.sub.y1, and is again re-phased with the
following positive gradient pulse G.sub.y2, so that an echo signal S.sub.1
is obtained from the FID signal S.sup.+.
Similarly, the gradient pulse G.sub.y3 together with the gradient pulse
G.sub.y4 operate to transform the half echo signal S.sup.- into a
complete echo signal S.sub.2.
Given flip angles grater than 50.degree., the relationship
S.sub.2 =S.sub.1 exp(-2T.sub.1 /T.sub.2)
is approximately valid for the signal amplitude of S.sub.2. This dependence
of the signal intensity can be used for calculating the transverse
relaxation time T.sub.2.
The acquired signals S.sub.1 and S.sub.2, which are phase-coded in the
x-direction by the phase-coding gradients G.sub.x1 and G.sub.x2 and are
frequency-coded in the y-direction by the gradient pulses G.sub.y2 and
G.sub.y3 serving as read-out gradients, are used for generation of an
image based on the two-dimensional Fourier transform method as described,
for example, in U.S. Pat. No. 4,070,611. Given a measuring time which is
substantially unaltered in comparison to conventional methods, the method
described above produces two images having different T.sub.2 contrast by
employing the signals S.sub.1 and S.sub.2. Depending upon the particular
application, the information content of one of the images can augment the
content of the other image, and thus provide increased diagnostic utility.
Another embodiment of the inventive method is shown in FIG. 5, wherein a
three-dimensional Fourier transform method is used, instead of the
above-described two-dimensional Fourier transform method. In this
embodiment, the slice selection gradient G.sub.z1, and the negative
auxiliary gradients G.sub.z2 and G.sub.z3, are omitted. Instead,
phase-coding gradients G.sub.z4 and G.sub.z5, which are inverted relative
to each other, are generated preceding and following each radio-frequency
pulse RF. The sequence remains otherwise unmodified. The three-dimensional
Fourier transform method, which is also disclosed in the aforementioned
U.S. Pat. No. 4,070,611, provides the advantage of an enhanced
signal-to-noise ratio, and the possibility of being subsequently able to
reconstruct every desired slice in the measuring volume.
Although modifications and changes may be suggested by those skilled in the
art it is the intention of the inventor to embody within the patent
warranted hereon all changes and modifications as reasonably and properly
come within the scope of their contribution to the art.
* * * * *
|
|
 |
|
 |
Description |
|
