Electricity: measuring and testing – Particle precession resonance – Using a nuclear resonance spectrometer system
Reexamination Certificate
2000-11-06
2003-03-04
Arana, Louis (Department: 2862)
Electricity: measuring and testing
Particle precession resonance
Using a nuclear resonance spectrometer system
C324S320000
Reexamination Certificate
active
06528999
ABSTRACT:
This application claims Paris Convention priority of DE 199 54 926.5 filed Nov. 16, 1999 the complete disclosure of which is hereby incorporated by reference.
BACKGROUND OF THE INVENTION
The invention concerns a method of correcting linear magnetic field inhomogeneities of a nearly homogeneous magnetic field B
0
in the investigation volume of a magnetic resonance apparatus, wherein magnetic resonance is excited in a sample, disposed in the investigation volume, through a radio frequency pulse, at least one additional linear magnetic gradient field is applied, and a magnetic resonance signal is measured.
A method of this type is known e.g. for a nuclear magnetic resonance tomography apparatus disclosed in U.S. Pat. No. 5,391,990.
In the known method, a bipolar gradient pulse train acts on a sample after radio frequency excitation, and a series of echoes is measured and stored in a signal matrix. The (temporal) position of the echo maximum and its displacement between the lines of the matrix are determined and used to calculate correction currents for linear shim coils.
The known method has i.a. the disadvantages that the chemical shift between water and fatty constituents of the signal produces undesired signal modulation and the method cannot be applied in an acceptable manner to samples having short T
1
/T*
2
times. In general, several iterations have to be carried out.
For this reason, there is a need for a rapid, uncomplicated direct shimming method which can be made insensitive to the influences of chemical shifts and can be successfully applied with samples having short relaxation times.
SUMMARY OF THE INVENTION
The object is achieved by a method of the above-mentioned type comprising the following steps:
An excitation radio frequency pulse is irradiated onto the sample;
A) a phase gradient G
ix
is applied in a predetermined direction x;
B) at a fixed time t
dx
after the radio frequency excitation pulse, a value S
ix
of the resonance signal coming from the sample is measured, digitized and stored;
C) the steps A) and B) are repeated several times with systematically altered strength of the phase gradient G
ix
;
D) the values of the measured resonance signals S
i
are compared and a maximum value S
xmax
is determined therefrom, to which a certain strength G
xmax
of the phase gradient is associated;
E) from the strength G
xmax
of the phase gradient determined in this fashion, a linear correction gradient magnetic field B(x) is determined for the predetermined direction x;
F) in subsequent measurements of magnetic resonance in the apparatus, the correction gradient magnetic field B(x) is applied to the investigation volume for homogenizing the magnetic field B
0
.
One single measuring point is recorded after a fixed predetermined time after each excitation rather than a complete signal echo. The time t
d
is always the same as is therefore the dephasing due to inhomogeneities of the magnetic field B
0
. Through application of a phase gradient in the interval between t=0 and t=t
d
, additional dephasing is produced which can be controlled in a defined manner. At the time of data recording, the two effects overlap. Through variation of the strength of the applied phase gradient, one obtains a maximum measuring signal when the dephasing influence of the field inhomogeneities in the respective direction is exactly compensated by the phase gradient. Only non-linear contributions and contributions perpendicular to the gradient direction remain.
One can minimize the influence of chemical shift in that the time t
d
is selected such that fatty and water contributions in the B
0
field are precisely in-phase at this relative point in time, in any case not of opposite phase.
The time t
d
can be selected largely freely, in particular, it can be very short for samples with short relaxation times T
1
/T*
2
.
Since the applied optimum phase gradient corresponds directly to the correction gradient, iterations are generally not required. Parameters which must be known in other methods, such as the position of the sample in the investigation volume, the resonance frequency or the applied RF performance, must not be exactly determined.
The method has been initially described with reference to one linear dimension but can be easily extended to several dimensions by carrying it out e.g. analogously for a further predetermined direction y which is preferably perpendicular to the direction x. It is not thereby necessary, however possible and preferred, if the above described correction for the x direction has already been carried out. In general, the effects simply add. Compensation in two dimensions is particularly favorable for examination of slices of an object. In multiple slice examinations, the field can be homogenized separately for each slice.
It is of course possible to extend the method in a corresponding manner to three-dimensional volumina by carrying it out analogously for a further predetermined direction z which is preferably perpendicular to the directions x and y. Therein, the field can be homogenized in the entire sample volume or, in connection with volume-selective measures, also for selected partial volumes and possibly for many different volumes within an object.
The easiest way of carrying out the method is to use the maximum measured signal of each respective S
ix
, S
iy
or S
iz
as the respective signal maximum S
xmax
, S
ymax
, S
zmax
.
Determination of a more exact, interpolated position is possible if each respective maximum of a smooth function S
x
(G
x
), S
y
(G
y
), S
z
(G
z
) is used as signal maximum S
xmax
, S
ymax
, S
zmax
which is fitted to the measured gradient strengths in the region of the maximum, preferably by a Gaussian function.
If the correction gradient field is to be determined very precisely and yet within a short time, it is recommended to initially carry out the method in rough gradient steps and then to repeat the steps A) to E) in the vicinity of the determined signal maxima, at least for one direction (x,y,z), thereby determining a refined value of the signal maxima.
A further possibility of increasing the accuracy consists in repeating, at least for one direction (x,y,z), the steps A) to E) for a different, preferably extended t
dx
, t
dy
or t
dz
.
Limitation to a narrow range about the already determined preliminary maximum is thereby possible. Longer t
d
means a longer dephasing time through the remaining inhomogeneities. Since one homogenization step has been carried out already, the signal strength is still sufficient.
The method is preferably a method of nuclear magnetic resonance and, in particular, is incorporated in a method of magnetic resonance imaging. Nuclear magnetic resonance imaging apparatus and, to an increasing extent, high-resolution nuclear magnetic resonance spectroscopy apparatus have gradient coils and shim systems as standard equipment. The method can be integrated into the existing software of, in particular, imaging apparatus without any hardware problems. In particular, it can be added to the actual measurement or be completely integrated therein such that the field is newly homogenized during the investigation program. As mentioned above, homogenization can be carried out individually for separate partial investigation areas.
In a preferred embodiment of the invention, the method is carried out for two preferably orthogonal directions, e.g. x,y and is preceded by a slice selection step, optionally including step A), which selects a slice in the investigation object perpendicular to a direction e.g. z. The selected slice can, for the general case, also be inclined to designated axes.
As an alternative, the method follows a previous volume selection step, optionally including step A), which selects a volume chosen from the investigation object.
With multiple volume experiments or multiple slice experiments, several sets of correction gradient fields are determined which are associated with different volumes or slices.
Correction currents can be determined from the determined correction gradi
Arana Louis
Bruker Medical GmbH
Vincent Paul
LandOfFree
Method of correcting linear field inhomogeneities in a... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method of correcting linear field inhomogeneities in a..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of correcting linear field inhomogeneities in a... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3044643