HiFi-MambaV2: Hierarchical Shared-Routed MoE for High-Fidelity MRI Reconstruction

TL;DR

HiFi-MambaV2 introduces a hierarchical shared-routed MoE architecture with frequency decomposition and content-adaptive computation, significantly improving high-fidelity MRI reconstruction from undersampled data by enhancing detail and structural coherence.

Contribution

The paper presents HiFi-MambaV2, a novel hierarchical shared-routed MoE model combining frequency decomposition with adaptive computation for superior MRI reconstruction.

Findings

Outperforms CNN, Transformer, and prior Mamba models in PSNR, SSIM, NMSE.

Achieves consistent high-frequency detail and structural fidelity improvements.

Demonstrates robustness across multiple datasets and acceleration factors.

Abstract

Reconstructing high-fidelity MR images from undersampled k-space data requires recovering high-frequency details while maintaining anatomical coherence. We present HiFi-MambaV2, a hierarchical shared-routed Mixture-of-Experts (MoE) Mamba architecture that couples frequency decomposition with content-adaptive computation. The model comprises two core components: (i) a separable frequency-consistent Laplacian pyramid (SF-Lap) that delivers alias-resistant, stable low- and high-frequency streams; and (ii) a hierarchical shared-routed MoE that performs per-pixel top-1 sparse dispatch to shared experts and local routers, enabling effective specialization with stable cross-depth behavior. A lightweight global context path is fused into an unrolled, data-consistency-regularized backbone to reinforce long-range reasoning and preserve anatomical coherence. Evaluated on fastMRI, CC359, ACDC, M4Raw, and Prostate158, HiFi-MambaV2 consistently outperforms CNN-, Transformer-, and prior Mamba-based baselines in PSNR, SSIM, and NMSE across single- and multi-coil settings and multiple acceleration factors, consistently surpassing consistent improvements in high-frequency detail and overall structural fidelity. These results demonstrate that HiFi-MambaV2 enables reliable and robust MRI reconstruction.