Image-fusion operators for 3D anatomical and functional analysis of the brain

The goal of the present work is to obtain images combining in a useful and precise way the information present in 3D volumes of medical imaging sets. We address two modalities combining anatomy (Magnetic Resonance Imaging or MRI) and functional information (Positron Emission Tomography or PET). Commercial imaging software often offers image fusion tools based on fixed blending or color-channel combination of different modalities, and various popular color Look-Up Tables (LUTs), without considering the anatomical and functional character of the image features. We used a sensible approach for image fusion, taking advantage mainly from the HSL (Hue, Saturation and Luminosity) color space, in order to enhance the fusion results. We further tested operators for gradient and contour extraction to enhance anatomical details, such as feature borders, plus other spatial-domain filters for functional features, which correspond to wide point-spread-function responses in PET images. A set of image-fusion operators was formulated and tested on brain images from PET and MRI acquisitions.