A Hybrid Multi-Modal Medical Image Fusion Framework Using Multi-Resolution and Multi-Scale Transforms with Block-Based Enhancement

Abstract

Medical image fusion is a critical process in clinical diagnostics, enabling the integration of complementary information from multiple imaging modalities into a single, more comprehensive image. This paper proposes a novel hybrid medical image fusion framework that leverages the combined advantages of multi-resolution and multi-scale analysis. The proposed method first generates an initial fused image by applying a multi-resolution technique, the Discrete Wavelet Transform (DWT), followed by a multi-scale technique, the Discrete Ripplet Transform (DRT). To further enhance the quality and preserve salient features from the source images, a block-based matching algorithm is subsequently applied to the initial fused image. This algorithm compares blocks of the fused image with corresponding blocks in the original Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) source images, selectively replacing them to construct a final, enhanced fused image. The performance of the proposed framework was evaluated against other transform-based methods, including DWT, DWT-DRT, and DWT-SVD-DRT, using two datasets of CT and MRI images. Quantitative analysis using metrics such as Peak Signal-to-Noise Ratio (PSNR), Mean Square Error (MSE), Structural Similarity Index Matrix (SSIM), and Entropy demonstrates the superiority of the proposed method. The experimental results show that our framework produces fused images with significantly improved visual quality and quantitative scores, making it a promising approach for enhancing diagnostic accuracy in clinical applications.