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

2026-03-17

AI Summary

This study addresses geometric distortions in echo-planar imaging (EPI) caused by magnetic susceptibility differences by proposing a fully automatic correction method grounded in optimal transport theory. The approach models the distortion field along the phase-encoding direction for each column of reversed EPI image pairs as a Wasserstein-2 barycentric displacement between intensity distributions and enhances smoothness by incorporating a bending energy regularization in the spectral domain. Innovatively, Morozov’s discrepancy principle is employed to enable self-regularization, automatically determining the optimal regularization strength without manual parameter tuning. Experiments on the Human Connectome Project (HCP) dataset demonstrate that the corrected EPI images achieve an average voxel-wise mutual information of 0.341 with T1-weighted structural images, outperforming FSL TOPUP (0.317), with a runtime of approximately 12 seconds on a single-core CPU.

Technology Category

• Search and Optimization: Non-convex Optimization
• Computer Vision: Learning & Optimization for CV
• Constraint Satisfaction and Optimization: Constraint Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

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.

Problem

• susceptibility distortion
• echo planar imaging
• geometric distortion
• EPI
• distortion correction

Innovation

• optimal transport
• susceptibility distortion correction
• parameter-free regularization
• Wasserstein barycenter
• echo planar imaging