Attention-aware non-rigid image registration for accelerated MR imaging

TL;DR

This paper presents an attention-aware deep learning framework for non-rigid MRI image registration that improves motion estimation accuracy at high acceleration factors, enhancing motion-compensated reconstruction quality.

Contribution

It introduces a novel transformer-based deep learning model combining local and global features for non-rigid MRI registration under high acceleration factors.

Findings

Reliable motion fields across different sampling trajectories and acceleration factors

Superior image quality in motion-compensated reconstruction compared to existing methods

Effective handling of artifacts caused by undersampling

Abstract

Accurate motion estimation at high acceleration factors enables rapid motion-compensated reconstruction in Magnetic Resonance Imaging (MRI) without compromising the diagnostic image quality. In this work, we introduce an attention-aware deep learning-based framework that can perform non-rigid pairwise registration for fully sampled and accelerated MRI. We extract local visual representations to build similarity maps between the registered image pairs at multiple resolution levels and additionally leverage long-range contextual information using a transformer-based module to alleviate ambiguities in the presence of artifacts caused by undersampling. We combine local and global dependencies to perform simultaneous coarse and fine motion estimation. The proposed method was evaluated on in-house acquired fully sampled and accelerated data of 101 patients and 62 healthy subjects undergoing cardiac and thoracic MRI. The impact of motion estimation accuracy on the downstream task of motion-compensated reconstruction was analyzed. We demonstrate that our model derives reliable and consistent motion fields across different sampling trajectories (Cartesian and radial) and acceleration factors of up to 16x for cardiac motion and 30x for respiratory motion and achieves superior image quality in motion-compensated reconstruction qualitatively and quantitatively compared to conventional and recent deep learning-based approaches. The code is publicly available at https://github.com/lab-midas/GMARAFT.