Quantification of T1 values by SNAPSHOT-FLASH NMR imaging

Abstract

SNAPSHOT-FLASH imaging has been described as a method of fast acquisition of NMR tomograms ( I ) . It is based on a slice-selective radiofrequency pulse of small flip angle (Y and the acquisition of magnetic field gradient echoes in a 2D Fourier imaging experiment. Under these conditions, the value of TR/ T, ( TR is the repetition time) can be extremely low. A whole 128 X 12%pixel tomogram may be taken in a fraction of a second. It has been shown that a SNAPSHOT-FLASH image is mainly spin-density-weighted. Contrasts with respect to NMR parameters can be introduced by the application of RF pulses before the whole image sequence ( I ) . For T1 measurements a sequence of several SNAPSHOT-FLASH tomograms is taken after the application of a 180” pulse with negligible delay between the pictures (about one millisecond). While an image is taken the longitudinal magnetization changes steadily. The intensity of the picture depends mainly on the longitudinal magnetization, which is present at the time when the amount of the phase-encoding gradient Gp is minimal. In a 2D Fourier imaging experiment Gp is switched symmetrically from a value of -GPO to +Gpo. Thus, the intensity of picture 12 after the 180” pulse depends on the longitudinal magnetization at the time t = ( IZ 1) r t 7/2, where T is the total acquisition time of a single tomogram. For T, = 1 s, the recovery in an inversion-recovery experiment will be attained after about five seconds. Hence, the whole inversion-recovery relaxation curve may be obtained by taking a sequence of at least eight pictures with 7 = 400 ms. However, one obtains an effective relaxation time Ty which is smaller than T1 because of the constant loss of longitudinal magnetization after each application of the (Y pulse. The exact evaluation of T, must be done with respect to this effect. The following treatment regards the “relative magnetization” m (total magnetization M devided by equilibrium magnetization MO). The T2 relaxation is neglected, since echo times as low as 1.4 ms can be used in SNAPSHOT-FLASH experiments. Before the pulse sequence used to measure an arbitrary gradient echo, there is longitudinal magnetization mj . After the cr pulse it is micos( a). During the measurement, there is spin-lattice relaxation leading to m(t) = 1 + [m(O) l]exp( -t/ T,). Hence, the longitudinal magnetization before the next pulse is mi+i = 1 t [micos(cu) l]exp( -T,/ T, ). In an inversion-recovery experiment the starting value is m. = 1. These conditions are fulfilled by