Model-Based Image Reconstruction for Dual-Energy X-Ray CT with Fast KVP Switching

Abstract

The most recent generation of X-ray CT systems can collect dual energy (DE) sinograms by rapidly switching the X-ray tube voltage between two levels for alternate projection views. This reduces motion artifacts in DE imaging, but yields sinograms that may be angularly under-sampled. This paper describes an iterative algorithm for statistical image reconstruction of material component images (e.g., soft tissue and bone) directly from such under-sampled DE data, without resorting to the interpolation operations required by conventional DE reconstruction methods.