Improving the image quality of spectral CT volume rendering

Spectral computed tomography (CT) is a new approach for medical diagnosis. Where traditional methods deliver a unique 3D volumetric dataset, spectral CT supplies multiple 3D volumetric datasets (energy bins). The novelty of this technique creates new opportunities and new challenges regarding visualization of the acquired data. In this paper we describe our approach and effort for providing high quality visualization of this new type of dataset. After a visual characterization of the data, we describe a set of image filtering algorithms we implemented and the results we obtained.