GraphSVR improves brain motion correction in diffusion MRI scans
GraphSVR: q-Space--Aware Graph-Based Slice-to-Volume Registration for Diffusion MRI
Computer Vision and Pattern Recognition
Summary
Diffusion MRI scans can get blurry or misaligned when people move during the test, especially for fast scans or kids who have trouble staying still. The authors created GraphSVR, a new way to fix these scans by treating the problem like a network graph that understands how different slices of the brain images connect in time and space. This method learns how to correct motion without needing any examples of perfect fixes. Their tests show it works better than current methods, especially when data is limited or motion is very bad.
What this means in practice
- •For medical imaging teams: Correct motion artifacts in diffusion MRI to produce clearer brain images in subjects who move during scanning.
- •For neuroimaging software developers: Incorporate graph-based motion correction into tools to better handle sparse data and severe motion in diffusion MRI pipelines.
Authors
Noga Kertes, Daphna Link Sourani, Alex M. Bronstein, Moti Freiman
Abstract
Diffusion-weighted imaging (DWI) remains highly vulnerable to subject motion, particularly in time-efficient protocols and in motion-prone populations. While slice-to-volume registration (SVR) can mitigate inter-slice and inter-stack misalignment, diffusion MRI introduces additional complexity due to diffusion-direction-dependent contrast and the requirement to align dozens of measurements within a common reference frame, effectively yielding a 4D registration problem. Existing approaches rely primarily on sequential modeling or pairwise similarity and often degrade under sparse gradient sampling or severe motion. We introduce GraphSVR, a q-space-aware graph-based framework for 4D SVR registration in DWI. GraphSVR represents slice groups as nodes in an acquisition-structured graph, with edges encoding temporal proximity, spatial slice geometry and diffusion encoding relationships. A graph neural network predicts globally consistent stack-wise rigid motion, optimized in a self-supervised, zero-shot manner using only an anatomical reference image, without requiring paired ground-truth motion. We evaluate GraphSVR using both fully synthetic diffusion simulations and realistic recombination-based simulations from real acquisitions with controllable motion severity and gradient sparsity. Performance is quantified using grid error (mm) and rotation error relative to known ground-truth transforms. Under severe motion, GraphSVR reduces grid error and rotation error by 73% compared to FSL eddy, the standard DWI motion-correction method, with the largest gains observed in sparse-direction regimes. These results demonstrate that explicitly modeling acquisition structure through graph-based reasoning improves robustness and global consistency in 4D DWI motion estimation. Code is available at https://github.com/nogakertes/GraphSVR.git.