FS-RWKV: Leveraging Frequency Spatial-Aware RWKV for 3T-to-7T MRI Translation

2025-10-09

AI Summary

To address the clinical limitation imposed by the scarcity of 7T MRI scanners, this paper proposes a 3T-to-7T MRI image synthesis method based on the RWKV architecture. Our approach introduces three key innovations: (1) the Frequency-Separated Offset-Shift (FSO-Shift) module, which enhances low-frequency global context while preserving high-frequency details; (2) the Structure-Fidelity Enhancement Block (SFEB), enabling frequency-aware feature fusion and adaptive structural enhancement; and (3) an integrated design combining discrete wavelet decomposition with omnidirectional spatial shifting to improve long-range dependency modeling and anatomical consistency. Evaluated on the UNC and BNU datasets across T1-weighted and T2-weighted modalities, our method consistently outperforms CNN-, Transformer-, GAN-, and state-of-the-art RWKV-based baselines in both quantitative metrics and qualitative assessment. It achieves superior anatomical fidelity and perceptual quality, setting a new state-of-the-art in 3T-to-7T MRI synthesis.

Technology Category

• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Computer Vision: Multi-modal Vision
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Ultra-high-field 7T MRI offers enhanced spatial resolution and tissue contrast that enables the detection of subtle pathological changes in neurological disorders. However, the limited availability of 7T scanners restricts widespread clinical adoption due to substantial infrastructure costs and technical demands. Computational approaches for synthesizing 7T-quality images from accessible 3T acquisitions present a viable solution to this accessibility challenge. Existing CNN approaches suffer from limited spatial coverage, while Transformer models demand excessive computational overhead. RWKV architectures offer an efficient alternative for global feature modeling in medical image synthesis, combining linear computational complexity with strong long-range dependency capture. Building on this foundation, we propose Frequency Spatial-RWKV (FS-RWKV), an RWKV-based framework for 3T-to-7T MRI translation. To better address the challenges of anatomical detail preservation and global tissue contrast recovery, FS-RWKV incorporates two key modules: (1) Frequency-Spatial Omnidirectional Shift (FSO-Shift), which performs discrete wavelet decomposition followed by omnidirectional spatial shifting on the low-frequency branch to enhance global contextual representation while preserving high-frequency anatomical details; and (2) Structural Fidelity Enhancement Block (SFEB), a module that adaptively reinforces anatomical structure through frequency-aware feature fusion. Comprehensive experiments on UNC and BNU datasets demonstrate that FS-RWKV consistently outperforms existing CNN-, Transformer-, GAN-, and RWKV-based baselines across both T1w and T2w modalities, achieving superior anatomical fidelity and perceptual quality.

Problem

• Translating 3T MRI to 7T MRI for enhanced image quality
• Overcoming computational limitations of CNNs and Transformers in synthesis
• Preserving anatomical details and global tissue contrast accurately

Innovation

• FS-RWKV uses frequency-spatial RWKV for MRI translation
• FSO-Shift module enhances global context via wavelet decomposition
• SFEB block reinforces anatomical structure through frequency fusion