The majority of sequences used in routine clinical magnetic resonance
imaging rely on the concepts involving the spin echo. Spin-echo sequen
ces require long acquisition times (1-10 minutes), but compared with f
aster gradient-recalled echo methods, spin-echo methods are relatively
immune to signal loss and distortions from field inhomogeneity and ti
ssue-induced susceptibility variations. Through modifications of inter
sequence repetition time (TR), echo formation interval (echo time [TE]
), and various gradient moments, image contrast can be altered to emph
asize tissue relaxation times T1, T2, or proton density. The TR and TE
values control the amount of T1 weighting and T2 weighting, respectiv
ely. At long TR intervals (approximately 10 x tissue T1 values) and mi
nimum TE values, the difference in signal intensity arising from relax
ation vanishes, and contrast arises solely from the differences in pro
ton density between the two tissues. Images formed with short TR inter
vals and long TE values exhibit very low signal-to-noise ratio and neg
ligible contrast and should be avoided. Recently, fast spin-echo seque
nces have partially overcome the limitation of long acquisition times,
with up to 16-fold reduction, by acquiring multiple lines in k space
with multiecho sequences.