SuCor: Susceptibility Distortion Correction via Parameter-Free and Self-Regularized Optimal Transport

TL;DR

SuCor is a novel susceptibility distortion correction method for EPI that uses optimal transport and automatic regularization, outperforming existing methods in accuracy and speed without manual tuning.

Contribution

It introduces a parameter-free, self-regularized optimal transport approach for susceptibility distortion correction in EPI imaging.

Findings

SuCor achieves higher mutual information with T1 images than FSL TOPUP.

SuCor runs in approximately 12 seconds on a single CPU core.

It automatically selects regularization strength without manual tuning.

Abstract

We present SuCor, a method for correcting susceptibility induced geometric distortions in echo planar imaging (EPI) using optimal transport (OT) along the phase encoding direction. Given a pair of reversed phase encoding EPI volumes, we model each column of the distortion field as a Wasserstein-2 barycentric displacement between the opposing-polarity intensity profiles. Regularization is performed in the spectral domain using a bending-energy penalty whose strength is selected automatically via the Morozov discrepancy principle, requiring no manual tuning. On a human connectome project (HCP) dataset with left-right/right-left b0 EPI pairs and a co-registered T1 structural reference, SuCor achieves a mean volumetric mutual information of 0.341 with the T1 image, compared to 0.317 for FSL TOPUP, while running in approximately 12 seconds on a single CPU core.