🤖 AI Summary
In diffusion MRI, echo-planar imaging (EPI) sequences are highly susceptible to eddy current–induced geometric distortions across diffusion-weighted volumes. Conventional image-registration–based correction methods (e.g., FSL Eddy) fail due to gradient-direction- and -strength–dependent signal attenuation and incur prohibitive computational cost. This paper introduces the first deep learning–based end-to-end eddy current distortion correction framework. We propose a dual-path CNN architecture that jointly performs image translation and deformable registration to directly estimate gradient-dependent deformation fields from minimal annotated data. Our method achieves correction accuracy comparable to FSL Eddy while accelerating inference by two orders of magnitude—enabling real-time preprocessing. By eliminating reliance on explicit physical modeling or iterative optimization, it establishes a new paradigm for efficient, robust, and clinically deployable diffusion MRI preprocessing.
📝 Abstract
Modern diffusion MRI sequences commonly acquire a large number of volumes with diffusion sensitization gradients of differing strengths or directions. Such sequences rely on echo-planar imaging (EPI) to achieve reasonable scan duration. However, EPI is vulnerable to off-resonance effects, leading to tissue susceptibility and eddy-current induced distortions. The latter is particularly problematic because it causes misalignment between volumes, disrupting downstream modelling and analysis. The essential correction of eddy distortions is typically done post-acquisition, with image registration. However, this is non-trivial because correspondence between volumes can be severely disrupted due to volume-specific signal attenuations induced by varying directions and strengths of the applied gradients. This challenge has been successfully addressed by the popular FSL~Eddy tool but at considerable computational cost. We propose an alternative approach, leveraging recent advances in image processing enabled by deep learning (DL). It consists of two convolutional neural networks: 1) An image translator to restore correspondence between images; 2) A registration model to align the translated images. Results demonstrate comparable distortion estimates to FSL~Eddy, while requiring only modest training sample sizes. This work, to the best of our knowledge, is the first to tackle this problem with deep learning. Together with recently developed DL-based susceptibility correction techniques, they pave the way for real-time preprocessing of diffusion MRI, facilitating its wider uptake in the clinic.