An assessment of eddy current sensitivity and correction in single-shot diffusion-weighted imaging.

Artefacts caused by eddy currents are a major problem in diffusion weighted imaging. This is particularly acute in experiments in which a number of images with differing degrees of diffusion weighting and/or differently oriented diffusion-weighting gradients need to be combined. The echo-planar imaging sequence is particularly sensitive to the effects of residual eddy currents, especially due to the low bandwidth in the phase-encoding direction. Two published schemes are investigated regarding the effectiveness of eddy current correction. That of Jezzard et al (1998 Magn. Reson. Med. 39 801-12) requires the acquisition of additional experimental data in order to perform a post-acquisition correction, whereas that of Wider et al (1994 J. Magn. Reson. A 108 255-8) attempts to reduce the eddy currents directly. It is found that the latter experiment gives a somewhat superior performance and a combination of the two approaches results in an almost complete elimination of artefact. An alternative single-shot imaging experiment to echo-planar imaging is given by sequences based on fast spin-echo methods, which should be insensitive to the effects of constant eddy currents. It is shown that the intrinsic eddy-current-related artefact level in such experiments is indeed low, residual artefacts being attributed to eddy current decay during the echo train. In situations of poor main magnetic field homogeneity or large eddy currents such sequences may be gainfully used instead of echo-planar imaging.