Flow effects in balanced steady state free precession imaging

Abstract

Fully balanced steady state free precession (SSFP) imaging (TrueFisp, FIESTA, Balanced FFE) has recently gained increased importance due to its high signal-to-noise ratio (SNR) and high blood-tissue contrast-to-noise ratio. These properties make it particularly useful for cardiovascular applications such as two-or three-dimensional cine MRI (1-28).Little has been reported, however, on the influence of flow on the formation of the SSFP signal and the subsequent steady state. In-flow of fresh, unsaturated spins can strongly influence the transition into the steady state and the steady-state signal itself or, in case of pulsatile flow, even result in periodically varying signal intensity that never reaches a steady state. As a result, signal intensities may be significantly altered and give rise to in-flow enhancement as compared to static structures of the same material and flow artifacts.The purpose of this article is to demonstrate the effects of in-flow on SSFP signal formation. Different-flow related-contributions to the SSFP signal were explored and used to simulate the effects of flow on SSFP signal intensities. The results of computer simulations were validated in phantom experiments using steady flow of different rates as well as pulsatile flow waveforms.The effects of imperfect excitation profiles and signal contributions from flowing "out-of-slice" spins had to be taken into account to fully understand and model the measured SSFP signal. The results show that SSFP imaging has considerable flow-related signal changes which strongly depend on off-resonance effects such as imperfect shim, eddy currents, or susceptibility effects. As a result, the excitation profile is replaced by a spatially nonuniform effective slice thickness that is flow-dependent. Implications for SSFP imaging in humans therefore include frequency offset-dependent in-flow enhancement with a concomitant sensitivity to flow artifacts. MATERIALS AND METHODS Theory and SimulationsThe temporal evolution of the magnetization (as a function of the number n of RF-excitations) is calculated using a matrix formalism similar to methods discussed by Hargreeves et al. (19) and Scheffler et al. (30). A train of equally spaced RF-pulses (constant TR) with constant relaxation times T 2 and T 1 was assumed for all simulations.Based on the Bloch equations, it can be shown that the magnetization M ជ n ϩ directly after the n th RF-pulse is given by the following recursive expression (see also Appendix):where R x (Ϯ␣) and D z (⌬ TR ) represent rotations of the magnetizations around x and z, respectively. R x (Ϯ␣) describes RF-excitation with flip angle ␣ and alternation of RF-phase which was realized by alternating the sign of ␣ for consecutive RF-excitations. D z (⌬ TR ) corresponds to a phase shift of the magnetization vector by ⌬ TR over the TR interval. E ជ 1 (TR,T 1 ) and E 2 (TR,T 2 ,T 1 ) represent T 1 -relaxation and T 2 -decay. The SSFP signal intensity M xy ϩ and phase ЄM xy ϩ immediately after the RF-pulse can be computed from the transverse com...

Non‐contrast‐enhanced vascular magnetic resonance imaging using flow‐dependent preparation with subtraction

“…In this study, such artifacts occurred only at or outside the edges of the acquired FoV. However, it is also known that the flow of blood through the banding artifact regions can lead to signal loss as the blood flows downstream, due to a disruption of the steady state (37). We postulate that this explains the slight reduction in vascular signal seen in some popliteal arteries.…”

On flow effects in balanced steady‐state free precession imaging: Pictorial description, parameter dependence, and clinical implications

“…(15). Other GRE sequences like GRASS or SPGR exhibit much faster signal decay once the RF excitation stops due to the large nonzero average gradient for these techniques.…”

Performance of steady‐state free precession for imaging carotid artery disease

“…However, its full capability for high-resolution carotid artery imaging under flowing conditions remains in doubt. In addition to suffering from intravoxel dephasing during the echo time, SSFP is also affected by center frequency offsets and magnetic field inhomogeneity, which may lead to improper refocusing between RF excitations (18,19). Moreover, flowing signal in SSFP can also produce out-of-slice refocused signal (18,20).…”