MR imaging of the liver at 1.5 T: value of signal averaging in suppressing motion artifacts

Abstract

Objective: Prior reports have suggested that signal averaging is not as effective as phase reordering for suppressing motion-related artifacts on T1-weighted spin-echo MR imaging of the liver at high field strengths. We hypothesized that with shorter TEs, signal averaging could be effective, and therefore undertook this study to compare signal averaging with phase reordering at 1.5T.

Subjects and methods: Thirty consecutive patients underwent MR imaging of the liver at 1.5 T with two T1-weighted spin-echo pulse sequences. In one sequence we used signal averaging and in the other we used phase reordering as the primary motion suppression techniques. For each patient, the order in which these sequences were performed was randomized; TR (400-500 msec), TE (11-12 msec), and field of view (32-34 cm) remained constant. Images were analyzed quantitatively for liver signal-to-noise ratio (SNR), liver-spleen contrast-to-noise ratio (CNR), and liver-lesion CNR (in 16 patients with focal liver lesions), as well as qualitatively (three observers, blinded to technique) for overall image quality.

Results: Signal averaging resulted in images with significantly greater liver SNR, liver-spleen CNR, and liver-lesion CNR than did phase reordering (p = .0001, .002, and .02, respectively). All observers showed a preference for the signal-averaged images. For one observer, this preference was statistically significant; for the others, it was not.

Conclusion: Signal averaging can be an effective means of suppressing motion-related artifacts on T1-weighted MR images of the liver obtained at 1.5 T.