Determination of Optimal Liver-Lesion Contrast in LOW-TIDE B-SSFP Imaging

Introduction: Previously, a variation of the Transition to Driven Equilibrium (TIDE) [1] called LOW-TIDE for Linear filter-based Optimal Window TIDE [2] approach to catalyzing the steady-state signal during continuous acquisition of image data in balanced SSFP had been presented. The LOW-TIDE technique has the advantage of smaller signal fluctuations at off-resonance frequencies for relatively long T2 species. In TIDE and LOW-TIDE scans, an (α/2)-(TR/2) preparation pulse is followed by a train of π (180°) pulses. This is followed by a smooth ramp down to the final flip angle. LOW-TIDE balanced SSFP is a potential alternative to breath-held T2-weighted single-shot turbo spin-echo imaging in body or abdominal applications. The LOW-TIDE approach provides T2 weighting for the typically (T2/T1) weighted b-SSFP sequence as well as intrinsic fat suppression when done with partial Fourier phase encoding. Here we investigate contrast characteristics for optimal LOW-TIDE imaging of liver metastases. Results from patient scans are used to validate theoretical predictions. Materials and Methods: A 4-term Blackman-Harris (B-H) window was used to calculate flip angles on the ramp down to the final flip angle. Simulations for evolution of magnetization at 1.5T were performed in Matlab® based on the algorithm in [3]. Liver T1 and T2 was assumed to be 46ms and 586ms [4] at 1.5T while T1 and T2 values for solid tumor were assumed to be 80ms and 1004ms [5], respectively. Magnetization typically follows pure T2 decay as long as the flip angle is maintained at 180°. Signal decay curves have a complex dependency on the number of π pulses and the duration of the ramp down pulses. Simulations were carried out for a range of asymptotic flip angles, # of π pulses and ramp down pulses. Behavior of spins at fat resonance (~217Hz at 1.5T) was also simulated. Adequate fat suppression dictates a limited range of effective echo times (and partial Fourier factor ⎯ PFF). Since liver-solid tumor contrast varies across k-space, a simple ramp weighting model for k-space [6] (with maximum weighting at center of k-space) can be used as a measure of effective contrast. Accordingly, contrast could be expressed as a cumulative sum: