Electricity: measuring and testing – Particle precession resonance – Using a nuclear resonance spectrometer system
Reexamination Certificate
2000-08-16
2003-01-28
Lefkowitz, Edward (Department: 2862)
Electricity: measuring and testing
Particle precession resonance
Using a nuclear resonance spectrometer system
C324S307000, C324S309000
Reexamination Certificate
active
06512372
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to an MRI (magnetic resonance imaging) apparatus, and more particularly to an MRI apparatus that can reduce ghost artifacts due to a Maxwell term phase error caused by a data acquisition read gradient.
FIG. 6
shows an exemplary pulse sequence in accordance with an EPI (echo planar imaging) technique.
In this pulse sequence, an excitation pulse RF
90
and a slice gradient SG
90
are applied. Subsequently, a phase encoding gradient P
n1
is applied. Next, an inversion RF pulse RF
180
and a slice gradient SG
180
are applied. Then, an alternating data acquisition read gradient r
1
, . . . , r
I
(I=4 in
FIG. 6
) having positive and negative polarities is consecutively applied, and at the same time, phase encoding gradients P
2
, . . . , P
I
are applied to sample 1
st
through I-th echoes e
1
-e
I
synchronously with sequential focusing of the 1
st
through I-th echoes, thereby collecting data d
n1
, . . . , d
nI
corresponding to the echoes e
1
-e
I
. This sequence is repeated for n=1, . . . , N, and data d
11
-d
NI
filling out a k-space are collected. This is referred to as an N-shot·I-echo process.
FIG. 7
is a schematic diagram illustrating a trajectory of collecting the data d
11
-d
NI
in a k-space KS, wherein N=4 and I=4.
When the k-space KS is divided into 1
st
through N·I-th rows (16 rows in
FIG. 6
) in the direction of a phase encoding axis, phase encodings P
n1
, P
2
, . . . , P
I
are applied so that data d
ni
for an (n+(i−1) N)-th row is collected by an i-th echo in an n-th shot.
Referring to
FIG. 8
, the k-space KS can be sequentially divided into a 1
st
echo block, which is filled with data d
n1
acquired from a 1
st
echo in each shot, through an I-th echo block (I=4 in FIG.
6
), which is filled with data d
ni
acquired from an I-th echo in each shot.
FIG. 9
is a diagram illustrating a phase error due to magnetic field inhomogeneity of a magnet.
The phase error due to magnetic field inhomogeneity increases in proportion to the time period from the excitation pulse RF
90
, as indicated by a magnetic field inhomogeneity phase error characteristic line.
If the time period from the excitation pulse RF
90
to the beginning of application of the data acquisition read gradient is the same among all the shots, all the data d
ni
corresponding to i-th echoes e
i
have a phase error magnitude of U
i
. Accordingly, the phase error exhibits a large stepped difference between the adjacent echo blocks, resulting in ghost artifacts.
Therefore, as shown in
FIG. 10
, the time period from the excitation pulse RF
90
to the beginning of application of the data acquisition read gradient is sequentially delayed from a 2
nd
shot to an N-th shot by a delay time of 1/N an echo space (an echo space=a time period between the adjacent echoes=a time width of a read gradient corresponding to one echo). This technique is referred to as echo shift. The magnitude of phase error now changes linearly in the direction of the phase encoding axis in the k-space KS and does not exhibit the large stepped difference in the phase error between the adjacent echo blocks. The ghost artifacts can thus be reduced.
FIG. 12
shows an exemplary pulse sequence in accordance with a GRASE (gradient and spin echo) technique.
In this pulse sequence, an excitation pulse RF
90
and a slice gradient SG
90
are applied. Subsequently, a read gradient r
0
is applied. Next, a j-th inversion RF pulse RF
180
_i (j=1, . . . , J. In
FIG. 12
, J=3) and a slice gradient SG
180
are applied. Then, an alternating data acquisition read gradient r
j1
, . . . , r
jI
(I=3 in
FIG. 12
) having positive and negative polarities is consecutively applied, and at the same time, phase encoding gradients p
j1
, . . . , p
jI
are applied to sample 1
st
through I-th echoes for the j-th inversion RF pulse e
j1
-e
jI
synchronously with sequential focusing of the 1
st
through I-th echoes, thereby collecting data d
nj1
, . . . , d
njI
corresponding to the echoes e
j1
-e
jI
. This sequence is repeated for n=1, . . . , N, and data d
111
-d
NJI
filling out a k-space are collected.
FIG. 13
is a schematic diagram illustrating a trajectory of collecting the data d
111
-d
NJI
in a k-space KS, wherein N=2, J=3 and I=3.
When the k-space KS is divided into 1
st
through N·J·I-th rows (18 rows in
FIG. 13
) in the direction of a phase encoding axis, a phase encoding p
ji
for an n-th shot is applied so that data d
nji
for an (n+(i−1) N+(i−1) N·J)-th row is collected by an i-th echo for a j-th inversion pulse in an n-th shot.
Also in the GRASE pulse sequence shown in
FIGS. 12 and 13
, ghost artifacts can be reduced by applying echo shift similarly to the EPI technique.
In the past, ghost artifacts due to a phase error caused by magnetic field inhomogeneity are thus reduced by the echo shift.
However, mere echo shift cannot fully reduce ghost artifacts because the conventional techniques have not accounted for ghost artifacts due to a phase error from a Maxwell term (which will be described in more detail later) caused by a data acquisition read gradient.
SUMMARY OF THE INVENTION
It is an object of the present invention to provide an MRI apparatus that can reduce ghost artifacts due to a Maxwell term phase error caused by a data acquisition read gradient.
In accordance with a first aspect of the invention, there is provided an MRI apparatus comprising pulse sequence generating means for generating a pulse sequence for collecting data, data collecting means for executing the generated pulse sequence to collect data, and image producing means for reconstructing an image from the collected data, wherein the pulse sequence generating means generates a pulse sequence for an n-th shot so that the following conditions are satisfied: (1) when a k-space is divided into 1
st
through N·I-th rows (wherein N and I are natural numbers not less than 2) in the direction of a phase encoding axis, repeating for N shots a pulse sequence which applies a data acquisition read gradient while inverting the gradient to focus I echoes per inversion RF pulse, and collecting data for filling out the k-space; and (2) appending an n-th Maxwell term correction read pulse before an inversion RF pulse in an n-th shot (wherein n=1−N), the n-th Maxwell term correction read pulse having a waveform whose time integral value is zero, and giving a bias phase error such that a Maxwell term phase error which is caused by the data acquisition read gradient and contained in the data filling out the k-space smoothly varies from the 1
st
row to the N·I-th row in the direction of the phase encoding axis (including, in a case of a shot taken as a reference, appending no Maxwell term correction read pulse only to that shot).
In other words, the present invention provides an MR imaging method for, when a k-space is divided into 1
st
through N·I-th rows (wherein N and I are natural numbers not less than 2) in the direction of a phase encoding axis, repeating for N shots a pulse sequence which applies a data acquisition read gradient while inverting the gradient to focus I echoes per inversion RF pulse, and collecting data for filling out the k-space, the method comprising the step of: appending an n-th Maxwell term correction read pulse before an inversion RF pulse in an n-th shot (wherein n=1−N), the n-th Maxwell term correction read pulse having a waveform whose time integral value is zero, and giving a bias phase error such that a Maxwell term phase error which is caused by the data acquisition read gradient and contained in the data filling out the k-space smoothly varies from the 1
st
row to the N·I-th row in the direction of the phase encoding axis (including, in a case of a shot taken as a reference, appending no Maxwell term correction read pulse only to that shot).
If a main magnet field is B
0
, and linear gradient magnetic fields in the X-, Y- and Z-direction
GE Yokogawa Medical Systems Limited
Kojima Moonray
Lefkowitz Edward
Vargas Dixomara
LandOfFree
MRI apparatus which has reduced ghost artifacts due to... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with MRI apparatus which has reduced ghost artifacts due to..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and MRI apparatus which has reduced ghost artifacts due to... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3047611