Multimodality Medical Image Fusion using Rotated Wavelet Transform

Abstract

This paper presents a novel approach to multimodality medical image fusion for better visualization of lesions and calcification. The algorithm utilizes source modalities as Computed Tomography (CT) and Magnetic Resonance Imaging (MRI). It is a feature based fusion technique in which Rotated Wavelet Transform (RWT) is used for extraction of edge-related features from both the source modalities. These features are used to create new frequency domain plane using maxima and entropy based fusion rules. The fusion process is useful in the analysis of the lesions for diagnosis, treatment, and post treatment reviews. The proposed technique is evaluated on the pilot study sets using objective analysis parameters like entropy, root means square error, edge quality measure, mean structural similarity index measure, etc. The fusion results of the proposed technique are compared with the existing fusion algorithms. The subjective analysis of the fused images by radiologists reveals that the fused images using RWT technique are superior and present all relevant anatomical structures.

Medical Image Fusion Using Integrated Guided Nonlinear Anisotropic Filtering with Image Statistics

“…Wavelet transform based fusion algorithms like discrete wavelet transform (DWT) [19,20], stationary wavelet transform (SWT) [21] have tremendous performance over the algorithms presented in the literature. Recent years, many extended versions of wavelet transform have done to improve the fusion performance further [22][23][24][25].…”