UniFuse: A Unified All-in-One Framework for Multi-Modal Medical Image Fusion Under Diverse Degradations and Misalignments

TL;DR

UniFuse is a comprehensive framework that jointly addresses multi-modal medical image fusion, misalignment, and degradation, outperforming existing methods by integrating restoration and fusion in a unified, adaptive approach.

Contribution

The paper introduces UniFuse, a novel unified framework that combines multi-modal image fusion, alignment, and restoration using a degradation-aware prompt and adaptive feature modules.

Findings

Outperforms existing methods on multiple datasets

Effectively handles misaligned and degraded images

Unifies alignment, restoration, and fusion in a single framework

Abstract

Current multimodal medical image fusion typically assumes that source images are of high quality and perfectly aligned at the pixel level. Its effectiveness heavily relies on these conditions and often deteriorates when handling misaligned or degraded medical images. To address this, we propose UniFuse, a general fusion framework. By embedding a degradation-aware prompt learning module, UniFuse seamlessly integrates multi-directional information from input images and correlates cross-modal alignment with restoration, enabling joint optimization of both tasks within a unified framework. Additionally, we design an Omni Unified Feature Representation scheme, which leverages Spatial Mamba to encode multi-directional features and mitigate modality differences in feature alignment. To enable simultaneous restoration and fusion within an All-in-One configuration, we propose a Universal Feature Restoration & Fusion module, incorporating the Adaptive LoRA Synergistic Network (ALSN) based on LoRA principles. By leveraging ALSN's adaptive feature representation along with degradation-type guidance, we enable joint restoration and fusion within a single-stage framework. Compared to staged approaches, UniFuse unifies alignment, restoration, and fusion within a single framework. Experimental results across multiple datasets demonstrate the method's effectiveness and significant advantages over existing approaches.