Frequency Error-Guided Under-sampling Optimization for Multi-Contrast MRI Reconstruction

2026-01-14

AI Summary

This work addresses key limitations in multi-contrast MRI reconstruction—namely, shallow reference fusion, insufficient exploitation of complementary information, and reliance on fixed undersampling patterns—by proposing a frequency error–guided joint optimization framework. The method introduces, for the first time, a conditional diffusion model to learn a prior over frequency-domain reconstruction errors and integrates this prior into a deep unfolding network, enabling end-to-end co-optimization of the undersampling pattern and reconstruction process. Additionally, it employs a spatial alignment and reference feature disentanglement strategy to effectively fuse complementary multi-contrast information. By synergistically combining model-driven and data-driven paradigms, the approach achieves superior reconstruction accuracy and physical interpretability across diverse imaging contrasts and sampling schemes, consistently outperforming state-of-the-art methods at acceleration rates ranging from 4× to 30×.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Magnetic resonance imaging (MRI) plays a vital role in clinical diagnostics, yet it remains hindered by long acquisition times and motion artifacts. Multi-contrast MRI reconstruction has emerged as a promising direction by leveraging complementary information from fully-sampled reference scans. However, existing approaches suffer from three major limitations: (1) superficial reference fusion strategies, such as simple concatenation, (2) insufficient utilization of the complementary information provided by the reference contrast, and (3) fixed under-sampling patterns. We propose an efficient and interpretable frequency error-guided reconstruction framework to tackle these issues. We first employ a conditional diffusion model to learn a Frequency Error Prior (FEP), which is then incorporated into a unified framework for jointly optimizing both the under-sampling pattern and the reconstruction network. The proposed reconstruction model employs a model-driven deep unfolding framework that jointly exploits frequency- and image-domain information. In addition, a spatial alignment module and a reference feature decomposition strategy are incorporated to improve reconstruction quality and bridge model-based optimization with data-driven learning for improved physical interpretability. Comprehensive validation across multiple imaging modalities, acceleration rates (4-30x), and sampling schemes demonstrates consistent superiority over state-of-the-art methods in both quantitative metrics and visual quality. All codes are available at https://github.com/fangxinming/JUF-MRI.

Problem

• multi-contrast MRI
• reconstruction
• under-sampling
• reference fusion
• complementary information

Innovation

• frequency error prior
• conditional diffusion model
• under-sampling optimization
• model-driven deep unfolding
• multi-contrast MRI