Spatial Unwarping
Echo-planar imaging (EPI) enables rapid functional imaging but introduces spatial distortion. Unwarping methods estimate and correct this distortion.
Multi-echo correction
Partial Fourier acquisition and in-plane acceleration can reduce EPI distortion, but may also reduce signal-to-noise ratio and do not eliminate distortion. Common correction methods use a separately acquired field map, as in FSL FUGUE, or images with reversed phase encoding, as in FSL TOPUP.
Van et al. (2026) introduced frame-wise Multi-Echo DIstortion Correction (MEDIC). MEDIC estimates field maps from phase data acquired as part of a multi-echo sequence. It requires no separate distortion-correction scan and estimates a field map for every volume, making the correction responsive to head motion.
The MEDIC authors provide the reference implementation in warpkit. niimath provides a compact, high-performance implementation through its -medic function.
Links
- Dymerska et al. describe Rapid Open-source Minimum spanning treE algOrithm (ROMEO) phase unwrapping.
- Van et al. (2026) introduce Multi-Echo DIstortion Correction (MEDIC).
- ROMEO.jl provides Julia source code under the permissive MIT license.
- warpkit provides C and Python code for MEDIC under a Washington University license that permits noncommercial use.