UEPS: Robust and Efficient MRI Reconstruction

TL;DR

UEPS is a novel MRI reconstruction model that improves robustness and efficiency by reconstructing coils independently and using physics-inspired designs, outperforming existing methods across diverse clinical shifts.

Contribution

The paper introduces UEPS, a new deep unrolled model that eliminates coil sensitivity map dependency and enhances robustness and efficiency for MRI reconstruction.

Findings

UEPS outperforms existing methods on diverse out-of-distribution tests.

UEPS achieves state-of-the-art robustness with low-latency inference.

The model demonstrates strong generalization across clinical shifts.

Abstract

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.