Multimodal Medical Image Fusion in NSCT Domain
Gaurav Bhatnagar, Zheng Liu, Q.M. Jonathan Wu · 2019
With the substantial proliferation in medical imaging modalities, multimodal medical image fusion has become a powerful tool for the clinical applications. The central idea is to collect the most pertinent information from different imaging modalities into a single output, which plays a critical role in efficient medical diagnosis. In this chapter, a novel multimodal medical image fusion framework is proposed in non-subsampled contourlet transform (NSCT) domain. The overall flow of the framework comprises of the following phases: Firstly, the source images are decomposed into low- and high-frequency sub-bands using NSCT. Secondly, phase congruency is adopted to deal with the low-frequency coefficients, and a new definition of directive contrast is proposed and is applied to fuse relevant information from the high-frequency coefficients. At the end, the fused image is constructed with the inverse NSCT transform on the processed low- and high-frequency coefficients. The superior performance of the proposed framework is finally evaluated and compared with that of existing approaches utilizing several clinical cases. Keywords : Medical Image Fusion, Non-Subsampled Contour Transform, Phase Congruency, Directive Contrast.