Mamba-Based Modality Disentanglement Network for Multi-Contrast MRI Reconstruction

TL;DR

This paper introduces MambaMDN, a novel dual-domain framework that leverages reference K-space data and modality disentanglement to improve multi-contrast MRI reconstruction, reducing artifacts and irrelevant information.

Contribution

The paper proposes a Mamba-based modality disentanglement network with iterative refinement for enhanced multi-contrast MRI reconstruction, addressing key limitations of existing methods.

Findings

Significantly outperforms existing multi-contrast MRI reconstruction methods.

Effectively utilizes K-space prior information to reduce aliasing artifacts.

Reduces irrelevant information contamination through modality disentanglement.

Abstract

Magnetic resonance imaging (MRI) is a cornerstone of modern clinical diagnosis, offering unparalleled soft-tissue contrast without ionizing radiation. However, prolonged scan times remain a major barrier to patient throughput and comfort. Existing accelerated MRI techniques often struggle with two key challenges: (1) failure to effectively utilize inherent K-space prior information, leading to persistent aliasing artifacts from zero-filled inputs; and (2) contamination of target reconstruction quality by irrelevant information when employing multi-contrast fusion strategies. To overcome these challenges, we present MambaMDN, a dual-domain framework for multi-contrast MRI reconstruction. Our approach first employs fully-sampled reference K-space data to complete the undersampled target data, generating structurally aligned but modality-mixed inputs. Subsequently, we develop a Mamba-based modality disentanglement network to extract and remove reference-specific features from the mixed representation. Furthermore, we introduce an iterative refinement mechanism to progressively enhance reconstruction accuracy through repeated feature purification. Extensive experiments demonstrate that MambaMDN can significantly outperform existing multi-contrast reconstruction methods.