Magnetization transfer contrast imaging using turbo spin echo protons of ''free'' water (pool A) and protons with restricted and continuous wave off-resonance irradiation was carried out on motions (pool B) in macromolecules and lipid bilayers ( 10). rat brain in vivo at 4.7 T. By systematically varying the off-reso-In the presence of such an interaction, a selective irradiation nance irradiation power and the offset-frequency, the signal intenof the ''immobile'' proton pool, at an offset frequency of sities obtained under steady-state for both transverse and longituseveral kilohertz with respect to the water resonance, will dinal magnetization were successfully analyzed with a simple cause a partial saturation of the former and a decrease in the binary spin -bath model taking into account a free water comintensity of the latter, thus creating contrast (11). The obpartment and a pool of protons with restricted motions bearing a served MT contrast can be modulated according to experisuper-Lorentzian lineshape. Due to important RF power deposimental constraints, which are the offset frequency of the MT tion, such experimental conditions are not practical for routine pulse along with its amplitude, duration, and shape. imaging on humans. An extension of the model was derived to describe the system for shorter off-resonance pulse duration, i.e.,
In clinical imaging, MT contrast is generated by many when the longitudinal magnetization of the free protons has not different acquisition protocols. Short (several milliseconds) reached a steady-state. Data sets obtained for three regions of inter- (1,5,(12)(13)(14) or long pulses (several hundreds of millisecest, namely the corpus callosum, the basal ganglia, and the temporal onds) (9) are used for irradiating the ''immobilized'' protons. lobe, were correctly interpreted for off-resonance pulse durations Considering the apparent T 1 values of free water pool in the varying from 0.3 to 3 s. The parameter sets obtained from the human at 0.5-2 T, its longitudinal magnetization is clearly calculations made it possible to predict the contrast between the not under steady-state. Thus it seems difficult to derive the different regions as a function of the pulse power, the offset frerelaxation parameters and the rate constant of exchange bequency, and pulse duration. Such an approach could be extended tween the two pools from such protocols. Nevertheless, such to contrast prediction for human brain at 1.5 T.