Functional and Anatomical Image Fusion based on Texture Energy Measures in NSST Domain

Abstract

The fused functional and anatomical image provides additional diagnostic information essential for precise diagnosis of disease and treatment planning. This work focused on obtaining a fused image that presents functional and anatomical information without loss and without distortion. We employed YIQ color model for functional information retention. The anatomical details are preserved through the use of texture energy measures in nonsubsampled shearlet transform (NSST) domain. The proposed method is tested on a dataset consisting of Single Photon Emission Computed Tomography (SPECT) and magnetic resonance imaging (MRI) images of brain tumor cases. Along with visual analysis, the fused images are quantitatively evaluated with five image fusion quality metrics.