Electricity: measuring and testing – Particle precession resonance – Using a nuclear resonance spectrometer system
Reexamination Certificate
1999-12-23
2002-04-02
Williams, Hezron (Department: 2862)
Electricity: measuring and testing
Particle precession resonance
Using a nuclear resonance spectrometer system
C324S307000, C324S300000
Reexamination Certificate
active
06366092
ABSTRACT:
BACKGROUND OF THE INVENTION
This invention relates to magnetic resonance imaging.
In magnetic resonance imaging, the subject to be imaged is positioned in a strong magnetic field, and magnetic resonant (MR) active nuclei such as protons in hydrogen atoms align parallel and anti-parallel to the magnetic field, precessing around the direction of the field at the Larmor frequency.
A transmit coil applies pulses of r.f. energy at the Larmor frequency in a direction orthogonal to the main field to excite precessing nuclei to resonance, which results in the net magnetisation of all MR active nuclei being flipped from the direction of the main magnetic field into a direction having a transverse component in which it can be detected by the use of a receive coil.
The received signal can be spatially encoded to produce two dimensional information about the distribution of MR active nuclei and hence of water and tissue.
Referring to
FIG. 1
, the main field is in the direction of the z-axis, and the strength of the magnetic field is made to vary along the direction of the z-axis by switching on z-gradient coils. The Larmor frequency of MR active nuclei will then also vary along the z-axis, and excitation can be confined to a slice such as slice A by appropriate choice of the bandwidth of the r.f. pulse.
Spatial information within the plane of the slice can be obtained by reading out the signals picked up by the receive coil in the presence of a gradient which varies e.g. in the y-direction, and by turning on and turning off a gradient in the orthogonal direction (in this case the x-direction) before the readout for a number of different increments of gradient strength and polarity in order to phase encode the received signal in that direction.
There are many occasions e.g. imaging the brain of a patient, when it is desired to produce information on several slices A, B, C etc. in which case the procedure outlined for slice A is repeated for slice B and then for slice C etc.
The signal picked up by the receive coil (the free induction decay—FID signal) could appear after as little as a very few ms (e.g. 2 to 3 ms) after the r.f. excitation pulse, and the collection of data corresponding to one r.f. pulse, including time for application of gradients and signal conditioning, could be as little as 5 to 6 ms. However, a desired repetition time (TR) at which the longitudinal and transverse components of magnetisation could have recovered to values necessary to produce a desired contrast could be 200 ms or much longer (up to a few seconds). Therefore, if TR were to be chosen as 40 ms, a FID signal at a first phase encode increment would be collected from slice A, and 40 ms later a second FID signal at a first phase encode increment would be collected from slice B. After five slices, an FID signal of a second phase encode increment would be collected from slice A.
In the case of a lower power magnet, a satisfactory signal-to-noise ratio might be produced only by completing the scanning procedure for all phase encode increments for all slices, and then repeating the data collection procedure a number of times.
To increase the signal-to-noise ratio without repetitions, or with a reduced number of repetitions, a technique is available for exciting a number of slices e.g. four, simultaneously, using only one r.f. pulse which behaves like four separate r.f. pulses of frequencies corresponding to the four slices in the presence of a3 magnetic field gradient, so that four times as much information can be collected after each r.f. pulse.
Of course, each FID signal collected would now relate to four slices, and the slices would have to be distinguished between in one way or another. One possible way would be to vary the r.f. phase so that e.g. on one excitation the r.f. pulse corresponding to one slice was in anti-phase compared to that corresponding to another. This set of data could then be identified from the signal after another excitation in which all were in the same phase. Permutating and combining such phase changes permits identification of the individual signals from the separate slices.
Another way of distinguishing between the slices would be to apply an additional slice encoding gradient. After each r.f. pulse, another gradient in the z-direction could be turned on and off before readout of the FID signals. For each phase encode gradient in the plane of the slices, a series of slice encode gradients would be applied over a range of gradient strengths and polarities.
While a useful amount of information is collected after each r.f. pulse in the interests of good signal-to-noise ratio, the time taken for such a data collection procedure increases.
The Applicants have appreciated that the time taken for data collection necessitated by slice encoding can be reduced by the application of a technique which has been proposed for phase encoding in the image plane.
This technique is referred to as Simultaneous Acquisition of Spatial Harmonics (SMASH): Fast Imaging with Radio Frequency Coil Arrays, Daniel K Sodickson, Warren J Manning, MRM 38:591-603 (1997) and WO-A-98/21600.
It will be recalled that the purpose of phase encoding e.g. in the x-direction in
FIG. 1
, is to have some means of distinguishing the contributions to the FID signal from precessing protons at various points along the x-axis. To do this, a gradient in the x-direction is turned on and turned off, creating a phase difference between the precessing protons at different points along the x-axis, before the readout takes place. A series of gradients are in fact applied, and the detected signals can be processed to sort out the locations of the precessing protons which contributed to the received signal. (The readout gradient resolves the contributions from the protons in the y-direction).
Referring to
FIG. 2
, consider for example a coil
1
(shown dotted) for producing a saggital (vertical longitudinal) section
2
through a spine
3
. In this case, the direction of the main magnetic field is along the z-axis, and the slices such as
2
are frequency selected by a magnetic field gradient along the x-axis. The slice
2
is spatially encoded by phase encoding in the z-direction and frequency encoding in the y-direction. A series of gradients in the z-direction are turned on and turned off before the FID signal is detected to accomplish phase encoding.
At each gradient, a phase difference will be generated between protons precessing at the same frequency but located at different positions along the length of the spine. Once this is done, it is possible to calculate the contribution to the FID signal from different positions along the length of the spine.
The SMASH technique refers to an array of overlapping coils
4
a,
4
b,
4
c,
4
d,
4
e,
4
f,
4
g,
4
h
instead of one large coil
1
. Each coil of the array now only views part of the region being imaged. The response patterns of the individual coils are shown in
FIG. 3
a.
If the outputs of the individual coils of the array are summed, the response is as shown in
FIG. 3
b
, and is similar to that of the conventional single coil. If the full amplitude output of the coil f is summed with the outputs of coils b, g, and e at reduced amplitudes, and the output of coil c and, with reduced amplitude, of coils b, d, and h are subtracted, the overall response of the coils will be such as to create a phase difference in the z-direction, as shown in
FIG. 4
b.
Thus, the FID signal will be different as regard to the contributions of protons precessing in phase located at the ends of the spine and the middle of the spine. The array corresponds to about one and a half cycles when combined in this way. If the array corresponded to one whole cycle, the response would be first order, etc.
The outputs of the coils of the array can be combined by alternately summing and subtracting them to produce a higher order response as shown in
FIG. 5
b.
The SMASH technique thus partially replaces gradient phase encoding in the plane of the slice
2
by a spatial encoding procedure that relies on the fac
Ehnholm Gosta J
Young Ian Robert
Fry John J.
Picker International Inc.
Shrivastav Brij B.
Williams Hezron
LandOfFree
Magnetic resonance imaging does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Magnetic resonance imaging, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Magnetic resonance imaging will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2882148