Magflelic Resona"ce Imaging. VoL 6, Ili l. 4OS - 4D , 1988
07JO-72SX/E8 S3 .00 + .00
Primed in th e USA. All righU reserved.
Copynghl © 1988 Perg:llmo n P ress pic
• Original Contribution
SOME FACTORS THAT INFLUENCE THE STEADY STATE
IN STEADY-STATE FREE PRECESSION
SAMUEL PATZ
Harvard Medical School and The Brigham and Women's Ho spit al, Departme nt of Radiology, Boston, MA 02115, USA
Steady-stilte free precess ion (SSFP) and particularly the spatiaJly periodic magnetization response with wavelength
A that resulfs in Ihe presence of an applied gradient is discussed. The maximum SSFP magnetization does not
always occur at the Larmor frequency but rather depends on both the phase cycling of the rf pulses and on the
rf tip angle. The slow flow sensitivity of SSFP also depends on A, Experimental data of the now sensitivity is
shown to be param eterized by a dimensionless dephasing parameter q, defined as the ratio of the distance
eled by a spin in one cycle of the SSFP sequence as compared 10 A. An example of the creation of a moving ref-
erence frame with the SSFP sequence is given. The mOdulation in image i.ntensity that results when the gradients
are ne3rly but not exactly compensated and), becomes large (i .e., tens of pixels long) is demonstrated. The pae)
length must be an integral uumber of ). 's in order to have a uniform image intensity from a uniform phantom .
Keywords: Steady-state free precession; SSFP; Flow; Moving reference frame; Image modulation .
INTRODUCTION
Stcady-state free pr ecession (SSFP) was first described
by Carr
l
in 1958. SSFP describes the driven equilib-
rium state that occurs upo n the application of a
strea m of rf pulses se parated by a time interval T and
where T « TI> T,. Under these conditions both a lon-
gitudinal and transverse magnetization steady state are
achieved. SSFP was on e of the first NMR techniques
to be used for imaging. In the early 1970s, Hin sha w
and Moore" and Hinshaw' showed how to use it to-
gether with three orthogon al oscillating gradients to
produ ce a point scanned image. Some of th e earliest
brain clinical images were produced using a projection
reconstru ction ve rsion of SSFP.
3
.
IO
The us e of SSFP
was sub stantially reduced when spin-echo techniques
that gave good graylwhite matter contrast were intro-
duced. Recently however, SSFP has enjoyed a renais-
sance with the deve lopment of a number of Fou ri er
ve rs ion s of SSFP .2.
4
. ll , 1)
SSFP is a rapid imaging technique with data acqui-
sition times in seconds rath er than minutes. Pa tient
studies with the Fourier SSFP technique of Hawkes
and P atz
4
were begun in October 1985 at Th e Brig-
ham and Women's Hospital, Boston, MA, and are de-
scribed in a companion article in this special is sue on
fast imaging.' The combination of rapid imaging and
good tissue contrast proved very attractiv e clinically.
Oth er Fourier SSFP techniques that have been intro-
duced offer variation s in the way the field refocused
echo is created and in the sli ce selection method. SSFP
has been shown 14 to be sensitive to slow flow on the
order of I mml s using gradient strengths readily avail-
able on clinical MRI mac hines (i.e., 0.25 G/ cm). The
ability to manipulate the tissue co ntra st with rf pulse
angle ha s been we Jl estabhshed."·7, 1.5 Furthermore,
Ma tsui et al
9
have shown that the T2 contrast can
depend on whether data are collected before or after
the rf pulse time.
This discussion reviews some basic SSFP concepts
a nd then describes the slow flow sensitivity of the
technique. A way in which to establish a moving ref-
erence frame with SSFP is treated next, together with
its app lication to flow measur ement. Th e particular
Fourier SSFP imaging pulse sequence used at our
institution is described; and finally some examples of
R ECErVED 9/ 8187; ACCEPTED 9/8187. im aging tec hniques. This work was s upported in pan by
Acknowledgments- I would like to thank S. Day for his grant number I ROI HL39035-01 fr om the Na ti onal Heart ,
assistance with the apparatus and measurements for the Lung. and Blood Institute of th e Department of HeaJth and
moving reference fra me experiments. I also thank R. C. Human Services.
Hawkes for many valuable discussions on both SSFP and
405