D2-LRR: A Dual-Decomposed MDLatLRR Approach for Medical Image Fusion

TL;DR

D2-LRR tackles the limitation of MDLatLRR by leveraging both base and detail features from LatLRR through learned projection matrices. It introduces a low-rank projection $P$ to extract base parts and uses the LatLRR-based detail extractor $L$ to obtain detail parts, enabling multi-level fusion with $f = f_b + \sum_{i=1}^n f_d^i$ where bases fuse via averaging and details via a nuclear-norm fusion. Trained on multi-modal medical data (CT, MRI, SPECT, PET) and evaluated on MRI–CT, MRI–SPECT, and MRI–PET pairs, D2-LRR demonstrates state-of-the-art performance across objective metrics and subjective quality. The approach shows robustness to sparse noise, improves structural preservation, and points to future integration with deep learning and registration techniques for broader clinical impact.

Abstract

In image fusion tasks, an ideal image decomposition method can bring better performance. MDLatLRR has done a great job in this aspect, but there is still exist some space for improvement. Considering that MDLatLRR focuses solely on the detailed parts (salient features) extracted from input images via latent low-rank representation (LatLRR), the basic parts (principal features) extracted by LatLRR are not fully utilized. Therefore, we introduced an enhanced multi-level decomposition method named dual-decomposed MDLatLRR (D2-LRR) which effectively analyzes and utilizes all image features extracted through LatLRR. Specifically, color images are converted into YUV color space and grayscale images, and the Y-channel and grayscale images are input into the trained parameters of LatLRR to obtain the detailed parts containing four rounds of decomposition and the basic parts. Subsequently, the basic parts are fused using an average strategy, while the detail part is fused using kernel norm operation. The fused image is ultimately transformed back into an RGB image, resulting in the final fusion output. We apply D2-LRR to medical image fusion tasks. The detailed parts are fused employing a nuclear-norm operation, while the basic parts are fused using an average strategy. Comparative analyses among existing methods showcase that our proposed approach attains cutting-edge fusion performance in both objective and subjective assessments.

Figures (8)

• Figure 1: MDLatLRR architecture.
• Figure 2: The reconstruction graph.
• Figure 3: D2-LRR framework.
• Figure 4: The proposed fusion method framework.
• Figure 5: There are three sets: the first row comprises CT, PET, and SPECT, while the second consists of MRI images.