Eddeep: a deep-learning framework for fast eddy-current distortion correction in diffusion MRI
Quick Answer
Eddeep is a deep-learning framework for rapid eddy-current distortion correction in diffusion MRI, achieving correction quality comparable to FSL Eddy while significantly reducing inference time.
Quick Take
Trained on UK Biobank data, it effectively addresses geometric distortions and head motion in a single forward pass, enhancing efficiency for large-scale studies and clinical applications.
Key Points
- Eddeep uses a two-stage approach: supervised image translation and unsupervised registration.
- Achieves correction quality comparable to FSL Eddy with reduced computational costs.
- Evaluated on both UK Biobank and Memodyn datasets for robustness.
- Significantly lowers inference time, supporting faster diffusion MRI processing.
- Code available for public access to facilitate further research.
DeepSignal Analysis
What happened
Eddeep is a deep-learning framework designed for rapid correction of eddy-current distortions in diffusion MRI. It achieves correction quality similar to FSL Eddy while significantly reducing inference time. The framework was trained on UK Biobank data and evaluated on both in-domain and out-of-domain datasets.
Key evidence
- Eddeep utilizes a two-stage approach: a supervised image translation network and an unsupervised registration network, enabling distortion correction in a single forward pass.
- The framework was trained on UK Biobank data and evaluated on datasets including Memodyn, demonstrating its effectiveness across different domains.
- Eddeep achieved correction quality comparable to FSL Eddy while substantially reducing inference time, indicating its potential for large-scale studies and clinical applications.
Why it matters
The development of Eddeep addresses a significant challenge in diffusion MRI, where eddy-current distortions can lead to misalignment and bias in microstructural analyses. By providing a faster and efficient correction method, Eddeep could enhance the processing pipelines for large-scale studies and improve clinical deployment of diffusion MRI techniques. This advancement may lead to more accurate imaging and analysis in medical research and practice.
Paper Resources
📖 Reader Mode
~2 min readAbstract:Diffusion MRI (dMRI) relies on diffusion-weighted echo-planar imaging, which is highly susceptible to eddy-current-induced geometric distortions. These distortions vary across diffusion volumes according to gradient strength and direction, causing between-volume misalignment that can bias downstream microstructural analyses. Current state-of-the-art correction methods, such as FSL Eddy, achieve high-quality correction through iterative prediction-correction schemes but are computationally expensive. We propose Eddeep, a deep-learning framework for fast eddy-current distortion correction in dMRI. Eddeep decomposes the problem into two stages. First, a supervised image translation network standardises the appearance of diffusion-weighted and b=0 images, removing contrast differences that hinder reliable registration. Second, an unsupervised registration network estimates both eddy-current distortion and between-volume head motion parameters under a physics-constrained quadratic distortion model, enabling correction in a single forward pass. The method was trained on UK Biobank data and evaluated on both in-domain (UK Biobank) and out-of-domain (Memodyn) datasets. Across a range of complementary metrics, including between-volume jitter, diffusion kurtosis imaging residuals, signal irregularity, and mutual information, Eddeep achieved correction quality comparable to that of FSL Eddy while substantially reducing inference time. These results demonstrate that deep learning can provide accurate and efficient eddy-current distortion correction without relying on iterative optimisation, supporting the development of faster diffusion MRI processing pipelines for large-scale studies and clinical deployment. The code is available at: this https URL.
| Comments: | Associated GitHub repo: this https URL |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2607.26292 [cs.CV] |
| (or arXiv:2607.26292v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2607.26292 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Antoine Legouhy [view email]
[v1]
Tue, 28 Jul 2026 21:33:50 UTC (4,496 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →ProMoE-FL: Prototype-conditioned Mixture of Experts for Multimodal Federated Learning with Missing Modalities
ProMoE-FL introduces a Prototype-conditioned Mixture-of-Experts framework for multimodal federated learning, effectively addressing missing modalities. It outperforms existing methods on four chest X-ray datasets, demonstrating superior feature synthesis capabilities in both homogeneous and heterogeneous settings.