Guided MRI Reconstruction via Schrödinger Bridge

2024-11-21

arXiv.org

AI Summary

To address the limited performance of multi-contrast MRI undersampled reconstruction caused by structural misalignment in cross-contrast prior fusion, this work introduces the Schrödinger Bridge (SB) into MRI reconstruction for the first time, establishing a differentiable diffusion bridge between the guidance and target image distributions. We propose I²SB-inversion—a novel inversion strategy inspired by image editing—that adaptively corrects cross-contrast structural discrepancies. By jointly incorporating multi-contrast guided sampling and data consistency constraints, our method achieves structure-aware, high-fidelity reconstruction. Evaluated on T1/T2-FLAIR paired datasets, it achieves up to 14.4× acceleration while significantly outperforming state-of-the-art diffusion-based and optimization-based methods in PSNR and SSIM. Moreover, the sampling process exhibits enhanced stability.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Imitation Learning & Inverse Reinforcement 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
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Web Mining and Content Analysis: Content-based information diffusion

Abstract

Magnetic Resonance Imaging (MRI) is a multi-contrast imaging technique in which different contrast images share similar structural information. However, conventional diffusion models struggle to effectively leverage this structural similarity. Recently, the Schr""odinger Bridge (SB), a nonlinear extension of the diffusion model, has been proposed to establish diffusion paths between any distributions, allowing the incorporation of guided priors. This study proposes an SB-based, multi-contrast image-guided reconstruction framework that establishes a diffusion bridge between the guiding and target image distributions. By using the guiding image along with data consistency during sampling, the target image is reconstructed more accurately. To better address structural differences between images, we introduce an inversion strategy from the field of image editing, termed $mathbf{I}^2$SB-inversion. Experiments on a paried T1 and T2-FLAIR datasets demonstrate that $mathbf{I}^2$SB-inversion achieve a high acceleration up to 14.4 and outperforms existing methods in terms of both reconstruction accuracy and stability.

Problem

• Improving MRI reconstruction from undersampled data using cross-contrast priors
• Addressing limitations of diffusion models in utilizing structural correspondence between contrasts
• Correcting inter-modality misalignment to reduce artifacts in guided MRI reconstruction

Innovation

• Schrödinger Bridge enables pixel-wise multi-contrast translation
• Inversion strategy corrects inter-modality misalignment artifacts
• Achieves high acceleration factor up to 14.4