Search papers, labs, and topics across Lattice.
This paper introduces UEPS, a deep unrolled model for MRI reconstruction designed to improve robustness to domain shifts by eliminating explicit coil sensitivity map (CSM) estimation. UEPS achieves this via an Unrolled Expanded design reconstructing each coil independently, progressive resolution refinement, and sparse attention tailored to MRI undersampling. Results on a large-scale zero-shot transfer benchmark show UEPS significantly outperforms existing methods across diverse out-of-distribution shifts while maintaining low-latency inference.
MRI reconstruction can be made dramatically more robust to clinical domain shifts by eliminating the need for explicit coil sensitivity map estimation.
Deep unrolled models (DUMs) have become the state of the art for accelerated MRI reconstruction, yet their robustness under domain shift remains a critical barrier to clinical adoption. In this work, we identify coil sensitivity map (CSM) estimation as the primary bottleneck limiting generalization. To address this, we propose UEPS, a novel DUM architecture featuring three key innovations: (i) an Unrolled Expanded (UE) design that eliminates CSM dependency by reconstructing each coil independently; (ii) progressive resolution, which leverages k-space-to-image mapping for efficient coarse-to-fine refinement; and (iii) sparse attention tailored to MRI's 1D undersampling nature. These physics-grounded designs enable simultaneous gains in robustness and computational efficiency. We construct a large-scale zero-shot transfer benchmark comprising 10 out-of-distribution test sets spanning diverse clinical shifts -- anatomy, view, contrast, vendor, field strength, and coil configurations. Extensive experiments demonstrate that UEPS consistently and substantially outperforms existing DUM, end-to-end, diffusion, and untrained methods across all OOD tests, achieving state-of-the-art robustness with low-latency inference suitable for real-time deployment.
