Phase insensitive preparation of single-shot rare for...

Details Phase insensitive preparation of single-shot rare for... Phase insensitive preparation of single-shot rare for...

2000-10-10

2002-12-03

Lefkowitz, Edward (Department: 2862)

Electricity: measuring and testing

Particle precession resonance

Using a nuclear resonance spectrometer system

C324S309000, C324S307000

Reexamination Certificate

active

06489766

ABSTRACT:

BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to diffusion imaging using a magnetic resonance imaging device and, more particularly, to a technique for modifying the RARE sequence to eliminate artifacts which relate to the strong sensitivity of multiple spin echo sequences to the phase of the prepared magnetization so that RARE may be used for diffusion and T
2
* imaging.
2. Description of the Prior Art
Recent advancements in scanner hardware and pulse sequence design have made possible single excitation images of excellent quality based on multiple spin echo sequences (MSES) such as RARE and GRASE. These sequences are an attractive alternative to echoplanar imaging because they suffer much less from chemical shift artifact and distortion than echoplanar. Echoplanar imaging is frequently used for forms of functional imaging in which a preparation sensitive to physiology is applied prior to the echoplanar readout. Such an approach can, in principle, also be used to prepare MSES.
Those preparation schemes that affect longitudinal magnetization, such as magnetization transfer saturation, arterial spin tagging, inversion recovery and saturation recovery, do not affect image quality and consequently work well with MSES. However, preparation schemes that alter transverse magnetization such as T
2
, diffusion, and T
2
* preparation can result in severely degraded images because they introduce unknown phase shifts in the transverse magnetization due to motion, chemical shift or magnetic field inhomogeneity. Most MSES rely on the amplitude stability of the Carr-Purcell-Meiboom-Gill (CPMG) sequence which depends upon the Meiboom-Gill (MG) phase condition so any changes in the phase of the transverse magnetization caused by preparation will result in rapid attenuation and modulation of the echo amplitudes causing signal loss and blurring in the images.
For T
2
preparation, phase errors will only occur if the timing of the sequence is inaccurate or subject motion in the presence of the relatively weak crusher gradients causes phase errors. Accurate timing and moderately cooperative subjects will probably be sufficient to obtain good quality T
2
prepared images. T
2
weighting can also be obtained by appropriate phase encode order.
On the other hand, T
2
* preparation will only be successful if all spins have the same chemical shift and the shim is outstanding. Though T
2
* weighted images will always exhibit signal loss in voxels where the static magnetic field gradient is very large, T
2
* prepared MSES images will also show severe signal loss or blurring in regions where the static magnetic field offset is large. This extra sensitivity to shim makes conventional T
2
* prepared MSES imaging unattractive.
Phase errors due to motion in the presence of large magnetic field gradients are the reason for the extreme motion sensitivity of multi-shot diffusion imaging. Single-shot echoplanar imaging is frequently used for diffusion imaging to avoid these motion induced errors. However, if a phase sensitive MSES is employed for diffusion imaging, severe signal loss and attenuation will occur.
The phase sensitivity of the CPMG sequence is clearly undesirable for these applications, so ways to avoid this sensitivity must be sought. The simplest approach is to ensure that the refocusing flip angle is exactly 180°. A MSES with exactly 180° refocusing pulses is completely insensitive to phase but very small deviations from 180° are sufficient to introduce artifacts. These artifacts are caused by the presence of multiple stimulated and spin echo pathways to produce signal contribution at the echo time. For magnetization satisfying the MG phase condition, these pathways add constructively and eventually achieve a temporary steady state echo amplitude. If magnetization is 90° from the MG phase, then the pathways interact destructively causing signal amplitude decay and oscillation. In most practical applications, including multi-slice imaging where the flip angle is not uniform across the slice, the flip angle cannot be made close enough to 180° to eliminate signals from these other pathways. In long echo train applications, such as single shot imaging, it is also desirable to lower the refocusing flip angle to minimize the power deposition in the subject. The favorable properties of reduced flip angle CPMG sequences have been described by Alsop in an article entitled “The Sensitivity of Low Flip Angle RARE Imaging, ”
Magn. Reson. Med.,
Vol 37, pp. 176-184 (1997) and by J. Hennig in an article entitled “Multiecho Imaging Sequences with Low Refocusing Flip Angles,”
J. Magn. Reson.,
Vol. 78, pp. 397-407 (1988).
C. S. Poon et al. in an article entitled “Practical T
2
Quantitation for Clinical Applications,”
JMRI,
Vol. 2, pp. 541-553 (1992) proposed crusher gradient schemes that can eliminate all but the primary refocused component. Unfortunately, these schemes require a large crusher amplitude that increases linearly with echo number. The added time required to apply the crusher gradients generally becomes unacceptable after only a handful of echoes. Several investigators have reported such sequences using crusher amplitudes that are too weak to fully dephase an individual voxel. These sequences employed non-selective refocusing pulses very close to 180° so the unwanted signal components are very weak. The quality of images obtained with slice selective or reduced flip angle refocusing pulses and these weaker crusher gradients would have to be evaluated. This spoiling approach has been employed to acquire. single-shot GRASE diffusion images. Because very few spin echoes and many gradient echoes were employed, the sensitivity to chemical shift and susceptibility artifacts was comparable to echoplanar. The highest quality GRASE images tend to employ many more radio frequency (RF) pulses and only a few gradient echoes. Spoiling of the CPMG sequence in this way will also cause the echo amplitudes to decrease rapidly with echo number if the refocusing pulse is reduced significantly from 180 degrees so reduction of the refocusing flip angle to lower power deposition is not possible.
A number of modifications to the CPMG sequence have been proposed to reduce errors in T
2
quantification or artifacts in multiple echo images by modulating the phase of the refocusing pulses. Some of these sequences can be interpreted as employing composite 180° pulses which are more insensitive to RF amplitude errors. Though these sequences work well for a few echoes when the flip angle is already near 180°, they begin to fail if the amplitude of the RF is reduced more significantly. They also usually increase the echo spacing and make single shot imaging more difficult. A two excitation method for producing phase insensitive images has also been proposed. Since the source of the phase uncertainty in diffusion imaging, motion, is not reproducible from excitation to excitation, the two shot method is not applicable.
An alternate approach to eliminating phase sensitivity of MSES sequences by crushing only some of the many stimulated and spin echo pathways has been presented by Norris et al. This approach was designed to overcome hardware limitations that precluded the precise timing and control necessary to achieve the MG phase condition. Modern clinical hardware can now readily achieve the MG condition because RARE has become an essential clinical tool. For the special applications of diffusion and T
2
* prepared RARE imaging, however, this approach is still very important. A limitation of the method is the large number of echoes which must be discarded before the signal is sufficiently stable to begin phase encoding.
A refinement of the Norris et al. method is desired that permits acquisition of data from the very first echo for acquisition of diffusion images. The present invention has been developed to meet this need in the art.
SUMMARY OF THE INVENTION
The present invention relates to a modification of Multiple Spin Echo Sequences (MSES), such as RARE, Fast Spin Echo, and GRASE, whi

Affiliated with

Also associated with

Rating

Phase insensitive preparation of single-shot rare for... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Phase insensitive preparation of single-shot rare for..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Phase insensitive preparation of single-shot rare for... will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2983818