U.S. patent application number 10/875134 was filed with the patent office on 2004-12-30 for method for the acquisition of moving objecs through nuclear magnetic resonance tomography.
Invention is credited to Hennig, Jurgen, Speck, Oliver.
Application Number | 20040263168 10/875134 |
Document ID | / |
Family ID | 33394963 |
Filed Date | 2004-12-30 |
United States Patent
Application |
20040263168 |
Kind Code |
A1 |
Hennig, Jurgen ; et
al. |
December 30, 2004 |
Method for the acquisition of moving objecs through nuclear
magnetic resonance tomography
Abstract
A nuclear magnetic resonance (NMR) tomography method for
investigating a target object, wherein radio frequency (RF) pulses
are irradiated into a target volume and/or RF pulses from the
target volume are detected, wherein the target volume is determined
by the frequency of the RF pulses and/or through magnetic field
gradients, and wherein the target object is moved relative to the
NMR tomograph during NMR data acquisition, is characterized in that
the frequency of the RF pulses and/or the magnetic field gradients
is/are changed during NMR data acquisition such that the target
volume covered by the RF pulses is moved relative to the NMR
tomograph at the same speed and direction of motion as the target
object during NMR data acquisition. This provides a method for
investigating a target object which moves relative to the NMR
tomograph during NMR data acquisition, which can be carried out in
a fast and simple manner.
Inventors: |
Hennig, Jurgen; (Freiburg,
DE) ; Speck, Oliver; (Freiburg, DE) |
Correspondence
Address: |
WALTER A. HACKLER, Ph.D.
ATTORNEY OF RECORD
SUITE B
2372 S.E. BRISTOL
NEWPORT BEACH
CA
92660-0755
US
|
Family ID: |
33394963 |
Appl. No.: |
10/875134 |
Filed: |
June 23, 2004 |
Current U.S.
Class: |
324/309 |
Current CPC
Class: |
G01R 33/56375 20130101;
G01R 33/3607 20130101; G01R 33/28 20130101 |
Class at
Publication: |
324/309 |
International
Class: |
G01V 003/00 |
Foreign Application Data
Date |
Code |
Application Number |
Jun 25, 2003 |
DE |
103 28 423.0 |
Claims
We claim:
1. Nuclear magnetic resonance (NMR) tomography method for
investigating a target object, wherein radio frequency (RF) pulses
are irradiated into a target volume and/or RF pulses from the
target volume are detected, wherein the target volume is determined
by the frequency of the RF pulses and/or through magnetic field
gradients, and wherein the target object is moved relative to the
NMR tomograph during NMR data acquisition, characterized in that
the frequency of the RF pulses and/or the magnetic field gradients
is/are changed during NMR data acquisition such that the target
volume covered by the RF pulses is moved at the same speed and in
the same direction of motion as the target object relative to the
NMR tomograph during NMR data acquisition.
2. Method according to claim 1, characterized in that during NMR
data acquisition, a slice-shaped target volume is investigated.
3. Method according to claim 2, characterized in that the target
object moves parallel to the surface normal of the slice-shaped
target volume.
4. Method according to claim 1, characterized in that the magnetic
field gradients are kept constant during NMR data acquisition and
only the frequency of the RF pulses is changed.
5. Method according to claim 4, characterized in that the following
applies for the Larmor frequency .omega. of the RF pulses:
.omega.(t)=.gamma.*B.sub.o+.gamma.*GS*v*t,wherein .gamma. is the
gyromagnetic ratio, B.sub.o is the static magnetic field, GS is the
magnetic field gradient, v is the speed of the target object and t
is the time.
6. Method according to claim 1, characterized in that the target
object is uniformly moved relative to the NMR tomograph.
7. Method according to claim 1, characterized in that the movement
of the magnetic field gradients is compensated for with respect to
the speed of the target object, in the direction of motion of the
target object, in particular, that the motion of the magnetic field
gradients is bipolarly compensated for.
8. Method according to claim 1, characterized in that a multi-echo
sequence in accordance with the principle of the RARE method is
applied, wherein the pulse phase is additionally adjusted to the
motion of the target object in accordance with the CPMG
conditions.
9. Method according to claim 1, characterized in that a multi-slice
technology is applied for investigating ns slices.
10. Method of NMR tomography for investigating a target object,
characterized in that several NMR data acquisitions are cyclically
repeated through methods in accordance with claim 1.
11. Method of NMR tomography for investigating a target object,
characterized in that several NMR data acquisitions are
subsequently carried out through methods in accordance with claim
1, and wherein after each NMR data acquisition, the position of the
target volume relative to the NMR tomograph is reset to an initial
position.
12. Method according to claim 10, characterized in that during an
NMR data acquisition, a slice-shaped target volume is investigated,
and the target object moves further by exactly one slice thickness
during one NMR data acquisition.
13. Method according to claim 10, wherein during NMR data
acquisition, a slice-shaped target volume is investigated,
characterized in that in NMR data acquisition with m individual
steps required for complete slice reconstruction after k=np*m/ns
individual steps, the position of the target volume is changed by
one slice thickness to obtain complete data for image
reconstruction after acquisition of N*m individual steps for a
total of N*ns-(ns/np-1) slices, wherein np>1, N>1.
14. Method in accordance with claim 1, characterized in that during
NMR data acquisition, two or more measuring sequences are applied
in a nested manner, wherein the measuring sequences generally
produce signals with different contrast, and wherein each measuring
sequence acts on a different partial volume of the target object.
Description
BACKGROUND OF THE INVENTION
[0001] The invention relates to a method of nuclear magnetic
resonance (NMR) tomography for investigating a target object,
wherein radio frequency (RF) pulses are irradiated into a target
volume and/or RF pulses from the target volume are detected,
wherein the target volume is determined by the frequency of the RF
pulses and/or through magnetic field gradients, and wherein the
target object is moved relative to the NMR tomograph during NMR
data acquisition.
[0002] NMR tomography methods with moving target objects are known
from Kruger et al., Magnet. Reson. Med. 47 (2002), 224, and also
from Scheffler, Proc. 9.sup.th Meeting ISMRM, Glasgow (2001),
1774.
[0003] NMR tomography is mainly used for medical diagnostics to
obtain information about the volume (i.e. the inside) of a target
object, in particular about diseased or possibly diseased regions
of a human body thereby utilizing the interaction between nuclei
and electromagnetic pulses.
[0004] NMR images are typically acquired in individual flat slices.
To investigate larger target objects, individual recordings of
several slices of the target object can be produced. When one slice
is completely recorded, the target object is displaced slightly
perpendicularly to the slice plane relative to the NMR tomograph
and a further slice is subsequently recorded. In these simple
cases, the target object is immobile during the actual NMR data
acquisition.
[0005] Recording methods of objects which are moved during the NMR
data acquisition (mainly for use for so-called total body imaging)
are known from literature (see e.g. Kruger et al, Magnet. Reson.
Med. 47 (2002), 224) which use different principles.
[0006] Methods were developed wherein the motion occurs within the
recording plane (i.e. the slice). With stationary magnetic field
gradients, such motions produce primarily displacement of the
recording data relative to the recorded target object. When
recording methods are used which are based on repeated recording of
in each case differently locally encoded signals, this displacement
differs in correspondence with the continuous motion of the object
from one recording step to the next. This displacement can be
compensated for through corresponding data post-processing when the
displacement speed is known.
[0007] Methods for recording images during motion perpendicular to
the recording plane are also known (see e.g Scheffler, Proc.
9.sup.th Meeting ISMRM, Glasgow, (2001), page 1774). In the
conventional methods, the displacement speed is thereby selected
such that the inconsistency of the data due to the displacement
remains small thereby hardly influencing the recording quality. In
general, these methods are carried out such that the motion advance
over an image acquisition is small compared to the thickness of the
selected volume (i.e. the volume to be investigated). Fast
acquisition techniques such as e.g. trueFISP or gradient echo
sequences are used for such recording methods, which are largely
stable compared to image artefacts caused by motion. These methods
still have the disadvantage that the small motion advance prolongs
the overall duration of the NMR investigation.
[0008] In contrast thereto, it is the underlying purpose of the
present invention to present a method for investigating a target
object, which moves relative to the NMR tomograph during NMR data
acquisition, which can be carried out in a fast and simple
fashion.
SUMMARY OF THE INVENTION
[0009] This object is achieved by a method of the above-mentioned
type in that the frequency of the RF pulses and/or the magnetic
field gradients is/are changed during the NMR data acquisition such
that during NMR data acquisition, the target volume covered by the
RF pulses is moved relative to the NMR tomograph at the same speed
and in the same direction of motion as the target object.
[0010] The target object is moved inside of the NMR tomograph. A
small part inside of the NMR tomograph is thereby selected as
target volume for data acquisition through the magnetic field
gradients, in particular slice selection gradients and the
frequency of the RF pulse. In accordance with the invention, the
frequency of the RF pulses and/or the strength of the magnetic
field gradients is/are changed during NMR data acquisition such
that the position of this target volume is carried along with the
moved target object thereby always obtaining the NMR data from the
same local region of the target object.
[0011] One variant of the inventive method is particularly
advantageous, wherein a slice-shaped target volume is investigated
during NMR data acquisition. Known slice selection gradients can
thereby be used to select the target volume. Carrying along of the
target volume together with the target object is moreover
facilitated when the target object moves perpendicularly to the
slice normal (i.e. the recording plane). The plane could also be
tilted as long as the motion has at least one component, other than
zero, perpendicular to the recording plane.
[0012] In a particularly preferred design of this method variant,
the target object moves parallel to the surface normal of the
slice-shaped target volume. Target objects with long extension in
one dimension can be guided through the inside of the NMR tomograph
(or also the NMR tomograph can be guided over the target object).
The target object (e.g. a person) can basically be investigated
along its full length, typically with several NMR data acquisitions
from different slices and continuous motion of the target object
relative to the NMR tomograph during and between all NMR data
acquisitions.
[0013] In a preferred variant of the inventive method, the magnetic
field gradients are kept constant during the NMR data acquisition
and only the frequency of the RF pulses is changed. Permanent
change of the magnetic field gradient with great precision is
difficult, in particular since the shieldings must be adjusted to
the varying gradient. In contrast thereto, exact adjustment of the
frequency of the RF pulses is relatively simple and high recording
quality of the NMR data acquisition can be obtained.
[0014] One particularly preferred design of the method variant is
characterized in that the following applies for the Larmor
frequency .omega. of the RF pulses:
.omega.(t)=.gamma.*B.sub.0+.gamma.*GS*v*t
[0015] wherein .gamma. is the gyromagnetic ratio, B.sub.0 is the
static magnetic field, GS is the magnetic field gradient, v is the
speed of the target object and t is the time. The target object
thereby moves at a constant speed relative to the NMR tomograph
perpendicularly to the recording plane. This frequency setting
ensures carrying along of the target volume with the target object
in a simple manner. Adjustment of linear temporal changes of the
Larmor frequency of the radio frequency (RF) pulses is easy.
[0016] In an advantageous method variant, the target object is
uniformly moved relative to the NMR tomograph thereby greatly
facilitating adjustment of the frequency of the RF pulses and/or
the magnetic field gradients. Uniform motion of the target object
can be easily set also in a fashion protecting the possibly human
target object. A uniform motion means a steady, straight
motion.
[0017] One method variant is also preferred, which is characterized
in that the motion of the magnetic field gradients is compensated
for in the direction of motion of the target object relative to the
speed of the target object, in particular in that the motion of the
magnetic field gradients is bipolarly compensated for such that
speed-dependent dephasing of the RF pulses is eliminated.
[0018] In a further preferred method variant, a multi-echo sequence
is used in accordance with the RARE method principle, wherein the
pulse phase is additionally adjusted in correspondence with the
CPMG conditions of the motion of the target object. The RARE method
utilizes several excitation pulses which precludes use of this
method for moving target objects without carrying along the target
volume as in accordance with the invention.
[0019] One method variant is also preferred, wherein a multi-slice
technology is used for investigating ns slices. The multi-slice
technology permits considerable reduction of the measuring time for
larger target objects.
[0020] One method of NMR tomography for investigating a target
object is also within the scope of the present invention, which is
characterized in that several NMR data acquisitions are cyclically
repeated by the above-mentioned inventive method. This permits
continuous investigation of target objects of any length in the
direction of motion. Each NMR data acquisition permits
investigation of another region of the target object which is
disposed in the measuring period in the inside of the NMR tomograph
in each case.
[0021] The present invention also includes a method of NMR
tomography for investigating a target volume which is characterized
in that several NMR data acquisitions are carried out successively
by the above-mentioned inventive methods, and wherein after each
NMR data acquisition, the position of the target volume relative to
the NMR tomograph is reset to an initial position. The NMR
tomograph has an admissible displacement region for the region of
the target object to be measured in which the NMR data is recorded.
After termination of the NMR data acquisition of this region
(typically this slice) the target volume returns to investigate a
new region of the target object. The target object can then be
effectively scanned region by region.
[0022] One variant of these two last-mentioned methods is
particularly preferred, wherein a slice-shaped target volume is
investigated during NMR data acquisition, and the target object
moves further by exactly one slice thickness during one NMR data
acquisition. In this case, the successively investigated regions of
the target object border directly, such that the target object can
be completely investigated (i.e. imaged) during the entire NMR
method.
[0023] One method variant is furthermore preferred, which is
characterized in that the position of the target volume is changed
by a distance ds during NMR data acquisition with m individual
steps required for complete slice reconstruction after k=np*m/ns
individual steps, which distance ds corresponds to the spatial
separation of the position of neighboring slices, such that after
acquisition of N*m individual steps for a total of N*ns-(ns/np-1)
slices, complete data is obtained for image reconstruction, wherein
np.gtoreq.1 (and preferably np=1), N.gtoreq.1. This saves time by a
factor of ns compared to the above-mentioned acquisition with the
individual slice method for investigating large target objects.
[0024] One variant of the inventive method is particularly
preferred, wherein during NMR data acquisition, two or more
measuring sequences are applied in a nested manner, wherein the
measuring sequences generally produce signals with different
contrast, and wherein each measuring sequence acts on a different
partial volume of the target object. Thus two or more image
contrasts can be acquired with one single passage of the target
object through the NMR tomograph.
[0025] Further advantages of the invention can be extracted from
the description and the drawing. The features mentioned above and
below may be used in accordance with the invention either
individually or collectively in arbitrary combination. The
embodiments shown and described are not to be understood as
exhaustive enumeration but have exemplary character for describing
the invention.
[0026] The invention is shown in the drawing and is explained in
more detail with reference to embodiments.
[0027] FIG. 1 shows a schematic illustration of the signal loss
through migration of the excited region of a moving target object
in accordance with prior art;
[0028] FIG. 2 shows a schematic illustration of the signal loss
through migration of a structure S of a moving target object in
accordance with prior art;
[0029] FIG. 3 shows a schematic illustration of the phase position
of a spin in a moving target object in accordance with prior
art;
[0030] FIG. 4 shows a schematic illustration of the inventive
method with individual steps A1 through Am;
[0031] FIG. 5 shows a schematic illustration of the inventive
principle of a continuous recording of a uniformly moving target
object;
[0032] FIG. 6 shows a schematic illustration of the inventive
principle of a continuous recording using a multi-slice method with
ns individual steps;
[0033] FIG. 7 shows a schematic illustration of an inventive NMR
data acquisition with 2 pulse sequences for different image
contrasts.
DESCRIPTION OF THE PREFERRED EMBODIMENT
[0034] The invention relates to NMR acquisition techniques, wherein
displacement occurs not in the plane of the recorded image (in
particular slice) but wherein the target object is displaced
perpendicularly to the image plane during recording.
[0035] For considering the effect of such a displacement, it must
be noted that data acquisition of a magnetic resonance (MR) image
is performed sequentially and in many acquisition methods through
repeated recording of individual steps, which differ through change
of the respective phase encoding of the individual steps in case of
the generally used Fourier transformation method.
[0036] The motion of the target object during the recording may
concern the signal intensity of each individual step, as well as
the consistency of the total number of individual steps used for
image reconstruction.
[0037] In the first case, signal loss occurs always when more than
one single, slice-selective radio frequency pulse is used to
generate a signal. This is the case e.g. for spin echo methods,
wherein an excitation pulse followed by at least one excitation
pulse, in the case of the RARE (also called TSE, FSE) method also
several excitation pulses, is/are used to generate a signal. If the
object moves during application of such a sequence, a signal loss
occurs already during recording of each individual step, because
part of the volume to be investigated moves out of the volume
detected by the measuring sequence thereby preventing measurement
of the spins contained therein (FIG. 1).
[0038] A further effect is obtained if a structure to be
represented remains to a large extent in the investigated slice
during recording of an individual step, but moves out of the volume
under investigation during recording of the total number of the
individual steps (i.e. the total duration of an NMR data
acquisition of one slice). Acquisition of the data required for
image reconstruction is incomplete for spins from such a structure
S within the moving target object, which produces image artefacts,
in particular unsharpness (FIG. 2).
[0039] In addition to these partial volume effects, further signal
losses or image artefacts occur in that the rephasing conditions
for the gradients used during acquisition for local encoding are no
longer met due to the motion of the spins (FIG. 3). The spins
experience dephasing during slice selection through a radio
frequency pulse in the presence of a slice selection gradient due
to their Larmor precession in the respectively local magnetic field
which results from superposition of the position-independent basic
field B0 with a position-dependent portion produced by the slice
selection gradient. Spins at different positions in the direction
of the gradient thereby experience different dephasing which must
be cancelled by corresponding compensating gradient steps before
signal readout. If the object does not move, this dephasing can be
compensated for in the simplest case through a following gradient
of negative amplitude and identical surface. If the object moves
during the acquisition, this motion produces a continuous change of
the magnetic field at the location of the spin. The phase
development of such moving spins corresponds then to a quadratic
form in correspondence with:
=.gamma..DELTA..PHI..intg.GS z(t)
dt=1/2v.sub.zt.sup.2.gamma.GS,
[0040] wherein .DELTA..PHI. is the phase shift, .gamma. is the
gyromagnetic ratio, GS is the slice-selection gradient, z is the
relative position of the region of the target object to be
investigated in the NMR tomograph, t is the time, v.sub.z is the
relative speed of the target object. This phase shift is relevant
mainly for so-called multi-echo methods (known as fast spin echo
(FSE), turbo spin echo (TSE) or RARE (rapid acquisition with
relaxation enhancement)) since they disturb the coherence of
refocusing over several echoes and therefore cause a drastic signal
loss.
[0041] On the basis of the above-mentioned considerations, the
following inventive measures are found to reproduce the consistency
of data acquisition. The slice defined by the slice selection
gradient and the frequency-selective excitation pulse is carried
along with the moving target object during acquisition such that
the pulse sequence acts in each case on identical volume regions of
the moving target object (FIG. 4). This is advantageously effected
in correspondence with the principle of slice selection in that the
slice selection gradient for the individual excitation steps
remains constant during application of the frequency-selective
radio frequency pulses--like for recording a stationary
object--while the selection frequency is changed in accordance with
the motion of the object.
[0042] In correspondence with the Larmor relationship, the
following applies:
.omega.=.gamma.B=.gamma.B.sub.0+.gamma.GS
z=.gamma.B.sub.O+.gamma.GS v.sub.zt=.omega..sub.0+.DELTA..omega.
[1]
[0043] wherein .omega. is the Larmor frequency, .gamma. is the
gyromagnetic ratio and B is the magnetic flux density at the
location of the spin. The latter is composed of the flux density
B.sub.0 of the magnet used and the contribution of the gradient GS
at the location z. The respective location of the moved spin at a
time t results thereby from the speed v.sub.z.
[0044] A slice selection gradient GS of a strength of 20 mT/m and a
displacement speed of 1 cm/s thereby produces a frequency shift of
.about.850 Hz over 100 ms.
[0045] For use in a multi-echo experiment, the frequency shift must
occur within each echo train, i.e. subsequent refocusing pulses
have different frequencies, wherein also the phase of the pulses
must be selected in accordance with the CPMG conditions such that
the phase coherence of the spins is maintained.
[0046] In the simplest case of using a pulse sequence which
inherently excites signals from one single slice and which requires
m recording steps for acquisition one complete data set for image
reconstruction, the acquisition is performed such that the
respectively selected excitation volume (target volume) is moved
over these m steps between the position A1 and Am in FIG. 5 in
correspondence with the displacement speed. The recording can
subsequently be repeated, wherein, when the same slice positions A1
through Am are selected in the moved target object, a new slice of
the target object which is correspondingly shifted parallel thereto
due to its motion, is recorded. If the speed of motion is thereby
selected such that Am is shifted relative to A1 by one slice
distance ds, through the cyclic repetition complete recording of
the object is obtained.
[0047] If multi-slice recording technology with ns individual
slices is used, the measuring principle can be modified as follows:
The slice position is carried along for k=np*m/ns recording steps,
wherein np>=1<=ns. After acquisition of the recording steps
A1 . . . Ak, the slice position in the resting coordinate system of
the analyzing magnet is switched further by one slice distance ds
(FIG. 6). After m recording steps, complete data for image
reconstruction from Ns=ns-(ns/np-1) slices is obtained, this cycle
is subsequently repeated N times to obtain in total complete data
for image reconstruction from (N-1)ns+NS=N ns-(ns/np-1) slices. N
is generally selected to be large such that the entire volume to be
investigated (e.g. the entire body) is large compared to the
recording volume covered by the ns slices. The fact that the data
from (ns-NS) slices is incomplete for image reconstruction, can
then be neglected.
[0048] Finally, it should be noted that the method can also be
applied using three-dimensional position encoding. The recording
steps A1 . . . Am thereby correspond to the individual steps for
recording a three-dimensional data set; the recording volume is
always carried along with the moved object.
[0049] A further, very essential and new type of application of the
measuring principle results from the finding that the method can be
modified such that recording of images with different contrast can
be carried out simultaneously through application of the described
principles at spatially separate positions which is shown in FIG.
7. The signals are thereby acquired in the positions in the NMR
tomograph marked by A1 . . . Am and B1 . . . Bm. Data acquisition
with completely independent measuring sequences is thereby carried
out within each of these two (or more) volumes under investigation
(target volumes), therefore producing images with completely
independent contrast. After passing the body once, (almost)
simultaneously generated data sets with different contrast can be
produced through continuous detection of the entire target object.
To vary the contrast behavior of the signals independently of each
other, the motion of the object must be sufficiently slow to ensure
that spins which enter the volume B under investigation have
sufficiently recovered from the previous signal excitation through
the sequence acting on A with respect to the contrast-relevant
parameters. The sequences acting on the partial target volumes can
thereby be configured absolutely independently of each other. To
obtain similar or identical volume coverages, it is in most cases
advantageous (but not absolutely necessary) that the sequences
attain similar or identical recording times for one recording
cycle. The recording can thereby be carried out either through
nesting of the individual steps (A1-B1-A2-B2 . . . ) or also--with
sufficiently fast recording techniques--through segment-wise
nesting up to nesting of the entire recording of a data set (A1-A2-
. . . Am-B1-B2- . . . Bm).
[0050] Recording methods for such nested recording with different
contrast, which are typically and frequently used in practice, are
combinations of frequently used sequences such as T1-weighted
recording, T2-weighted recording, STIR, FLAIR, diffusion-weighted
recordings and many more.
[0051] The speed-dependent dephasing of the signals which occurs
during the measurement due to the change of location of the target
object and therefore of the spins to be investigated, can be
eliminated using gradients which are compensated for with respect
to a constant speed. This is analogous to the known principle of
flux-compensated measuring methods (see e.g. Duerk et al, Magn.
Reson. Imaging 8 (1990), 535). A necessary and sufficient condition
for this is that, in addition to the integral under the gradient
between excitation and reading, also the integral of the square of
the gradient is set to zero. This is obtained in the most simple
case through so-called bipolar motion-compensated gradients.
[0052] Explanation of the Figures:
[0053] FIG. 1 shows the signal loss in a recording with fixed
location of a partial volume A, wherein the object O moves in the
direction of the arrow. If the partial volume A detected by the
excitation pulse is identical with the partial volume B detected by
the refocusing pulse, part of the originally excited spins (area C
with inclined hatching) has moved out of the slice (i.e. the target
volume) before refocusing and does therefore not contribute to
signaling and the observed signal therefore originates only from
the partial volume D (crosshatched).
[0054] FIG. 2 shows the same for a structure S which moves out of
the partial volume A detected by the slice-selective acquisition
(i.e. the target volume) during acquisition of all individual steps
used for image reconstruction and reaches the position S' outside
of this volume A under investigation at the end of image
acquisition. Since this structure provides no signal contribution
during part of data acquisition, the data amount required for
spatial encoding is incomplete which produces image artefacts.
[0055] FIG. 3 shows the effect of the motion of the object on the
phase .PHI.(t) of spins which move along a gradient GS during the
acquisition. B is thereby the magnetic flux density which changes
through application of a gradient linearly along the direction z of
the gradient. RF shows the excitation pulse of an MR sequence,
GS(z) shows the schematic diagram for the gradient used. A signal
for stationary spins is produced when the integral over the
gradient (hatched area) is zero. This condition is not met for
moving spins; they experience dephasing .DELTA..PHI.(v) which is
proportional to the speed of motion.
[0056] FIG. 4 shows the principle of the inventive method: the
position of the respectively investigated slice (i.e. the target
volume) is carried along with the moved target object in the NMR
tomograph, such that the positions of the individual steps A1 . . .
Am differ in space relative to the NMR tomograph, but detect
respectively identical volumes of the moved target object. For spin
echo methods, wherein several pulses are used for each individual
step, the slice position is maintained also within each echo train.
The respective slice positions are shown next to each other only
for reasons of clarity. The slice continuation will generally be
small compared to the slice thickness in dependence on the speed of
motion.
[0057] FIG. 5 shows the principle of a continuous acquisition of a
target object moving at a speed v.sub.z. During the time t0 . . .
t1, the individual steps A1 . . . Am which are required for
acquisition of a data set from the target volume 1 are recorded
wherein the recording volume is moved together with the target
object. The procedure is subsequently repeated with a second
neighboring volume 2 within a time t2 . . . t3. With multiple
repetition, a target object of any size can be successively and
continuously investigated.
[0058] FIG. 6 shows the principle of continuous recording of a
volume with application of a multi-slice acquisition method with ns
individual slices: The recording slice is carried along with the
moved object over k=m/ns individual steps, and the recording slice
is subsequently displaced in the object by exactly one slice
position ds.
[0059] FIG. 7 shows the principle of continuous acquisition of data
with sequences which effect different image contrasts: in the
partial volume characterized by A1 . . . Am, a sequence with a
certain contrast behavior is carried out in accordance with the
above principles. In B1 . . . Bm, a second and different sequence
is applied in a nested manner thereto, which produces a different
contrast. The object is detected in two different contrasts with
continuous acquisition through one single passage due to the motion
of the object.
[0060] The invention presents an MRT method, wherein the object
moves transversely to the direction of the selected volume during
data acquisition, and wherein the signal losses caused through
motion of the object and inconsistencies of recorded data are
prevented in that the volume to be investigated is carried along
with the moving object and all general conditions required for
signal generation with respect to magnetic field gradient and the
radio frequency pulses used, are adjusted to the motion of the
object.
* * * * *