Sliding frequency steady-state precession imaging

Details Sliding frequency steady-state precession imaging Sliding frequency steady-state precession imaging

2001-11-20

2003-09-23

Arana, Louis (Department: 2859)

Electricity: measuring and testing

Particle precession resonance

Using a nuclear resonance spectrometer system

C324S309000

Reexamination Certificate

active

06624630

ABSTRACT:

BACKGROUND OF THE INVENTION
The present invention relates to the magnetic resonance imaging arts and particularly to steady-state free precession (SSFP) imaging.
In steady-state free precession imaging, a series of magnetic resonance excitation pulses are applied at relatively short intervals, typically in the 2-15 millisecond range. After this series of pulses is applied for about three times the T
1
relaxation time (between one and six seconds for many common tissue types), the magnetic resonance is stabilized in a steady-state condition. Thereafter, following each magnetic resonance excitation pulse, a phase-encoding gradient and a read gradient are applied to read out one phase-encoded data line until a full set of data lines are collected. Conventional steady-state free precession imaging sequences take advantage of signal generation along the longitudinal axis or both the longitudinal axis and the transverse plane.
Area and moment balancing of gradient pulses which are applied in addition to frequency unwinding pulses which maintain coherence of the spin system rotating frame of reference have been used to increase image signal-to-noise ratios by making efficient use of the superposition of the rising and falling components of the resonance signal. Balanced gradient moments and short repetition times make steady-state free precession sequences insensitive to many types of patient motion and allow spins of flowing blood or cerebral fluid to come to a steady-state.
Strong tissue contrast is obtained on the basis of the T
1
and T
2
relaxation times and the tip angle. Low tip angle images have a T
1
or proton density weighted contrast. High tip angle images have a contrast that is more T
2
dependent. At the Ernst angle, the image contrast is relatively independent of the repetition time and becomes proportional to the ratio of T
1
and T
2
of the imaged tissue.
Completely balanced sequences possess significant disadvantages. The pulse train produces a spectrally varying periodic excitation pattern with a frequency period proportional to the reciprocal of the sequence repetition time. This excitation pattern has two distinct null points per period that result in image banding artifacts. In order to achieve a stable steady-state in both the longitudinal and transverse magnetization, the excitation pulses are applied for several T
1
before the signal is sampled. Often, the stabilization time is a significant fraction of the total imaging time, often more than half.
The periodic nature of the excitation pattern leads to banding artifacts that tend to follow the magnetic field contours in steady-state free precession images, because magnetic field inhomogeneity contributes to the local spectral offset in each imaging pixel. In general, the appearance of banding artifacts is governed by the total phase precession gained by spins during the interpulse (TR) period. Common sources of phase precession in liquid state imaging include: resonance offset, chemical shift, magnetic field inhomogeneity, spin coupling, unbalanced gradient pulses, and unbalanced changes in the transmitter carrier frequency and receiver demodulation frequency or their phases. An effective (average) spectral offset (radians/sec) can be defined for each spin species in any pixel by dividing its total phase gain between pulses (radians), by the interpulse spacing (seconds).
One proposed solution for banding artifacts is the use of very short repetition times to reduce the total phase gain from local changes in spectral offset. Another solution is to average a plurality of excitations or images with a range of spectral offsets such that the banding artifacts average out or cancel, because of the periodic nature of the steady-state spectral response. Both solutions, however, have drawbacks.
Ultra-short repetition times compact the time available for phase-encode and read gradients. As the times become shorter, the gradients need to turn on and off more quickly, i.e., a higher slew rate. Typically, higher strength gradients are used to offset the shorter read times. Such systems are expensive and may be unfavorable to patient safety and comfort. Rapidly changing magnetic fields can induce both muscular and neural stimulation. The use of strong gradients also requires the use of wide bandwidth receivers to digitize the read out resonance signals. Such wide bandwidth receivers decrease the signal-to-noise ratio counteracting a primary advantage of the steady-state free precession method. Further, short repetition times severely restrict image resolution.
Averaging spectral offset magnetic resonance signals or complex valued images created from such signals successfully cancels banding artifacts. The RF pulses in the steady-state free precession excitation train are shifted in constant phase increments, e.g., 90°, for each successive RF pulse. A similar phase shift is also applied to the signal detected in the imager's receiver. These combined phase shifts change the frequency of the rotating reference frame. Hence, a 90° phase increment moves the spectral excitation pattern by a frequency offset equal to ¼ of a period. A 180° phase increment shifts the pattern by ½ of a period, and so on. Collecting a set of data lines for each of a plurality of different phase cycles. with different phase increments spaced equally over any 360° period allows averaging to cancel the banding artifacts. Alternately, a Fourier transform applied across the spatially encoded data dimension can be used to average or combine the phase shifts. The collection of the steady-state free precession signal with a series of increment phase cycles is the equivalent to encoding the data with spectral offset in addition to the usual spatial encodings that are accomplished by the gradient pulses. Calculation of a mean or averaged signal is equivalent to the calculation of the central (k=0) Fourier coefficient for the spectral domain.
A primary drawback of the signal averaging or transform approach is that a large number of averages are collected to cancel the banding artifacts, typically at least four. Time is allocated for the system to regain its steady-state after each change of the phase cycle. This causes image data to be averaged in a serial fashion rather than in a sequential fashion. That is the four or other plural number of like data line sets with each of the averaged spectral offsets are collected at temporally different times rather than sequentially.
The present invention provides a new and improved steady-state free precession imaging technique that overcomes the above-referenced problems and others.
SUMMARY OF THE INVENTION
In accordance with one aspect of the present invention, an improvement is provided in steady-state free precession magnetic resonance imaging sequences in which a plurality of sets of data lines are collected with different spectral offsets. The improvement includes changing the spectral offset of the RF excitation as each data line is collected to modulate, but not destroy the steady-state response of the spin system.
In one aspect of the invention, the spectral offset is changed sufficiently slowly that steady-state condition can be reversibly modulated and anneal between the acquisition of each data line.
In accordance with another aspect of the present invention, the spectral offset is cycled repeatedly through a finite series of states to generate a periodically repeating cycle of spectrally shifted steady-state responses.
In accordance with another aspect of the present invention, a method of magnetic resonance imaging is provided. A region of interest of a subject to be imaged is placed in an examination region in a main magnetic field. Resonance is repeatedly excited with a RF pulse at regular intervals. After steady-state free precession has been established, RF pulses continue to be applied and at least phase encode and read gradients are applied following each RF pulse to generate a set of data lines in k-space. During the generation of the set of data lines, the phase of th

Affiliated with

Also associated with

Rating

Sliding frequency steady-state precession 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 Sliding frequency steady-state precession imaging, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Sliding frequency steady-state precession imaging will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3066491