MIF-SWT-TVM: A Hybrid Multi-Modal Medical Image Fusion Using Stationary Wavelet Transform and Gradient-Based Total Variation Minimization

The Multi-Modal Medical Image Fusion (M3IF) plays a critical task in combining corresponding important information from different imaging modalities (e.g., Magnetic Resonance Imaging (MRI), Computed Tomography (CT), Positron Emission Tomography (PET) etc. to improve clinical diagnosis and decision-making. This paper proposes a hybrid image fusion technique (MIF-SWT- TVM) that leverages Stationary Wavelet Transform (SWT) and Total Variation (TV) Minimization to enhance the visual quality and diagnostic accuracy of fused medical images. The method fuses approximation coefficients (ACs) using an averaging strategy and detail coefficients (DCs) using an absolute maximum selection strategy. The TV regularization is applied post-fusion to reduce noise, suppress artifacts, and enhance structural continuity, leading to a more refined and clearer fused image. The MIF-SWT-TVM is compared with some of the traditional wavelet-based and other fusion techniques, demonstrating superior preservation of anatomical structures and contrast enhancement. The MIF-SWT- TVM effectively enhances diagnostic features, reduces noise, and provides better visual quality compared to existing fusion techniques.