A Two-Step Framework for Multi-Material Decomposition of Dual Energy Computed Tomography from Projection Domain

Abstract

Background and Purpose: Dual-energy computed tomography (DECT) utilizes separate X-ray energy spectra to improve multi-material decomposition (MMD) for various diagnostic applications. However accurate decomposing more than two types of material remains challenging using conventional methods. Deep learning (DL) methods have shown promise to improve the MMD performance, but typical approaches of conducing DL-MMD in the image domain fail to fully utilize projection information or under iterative setup are computationally inefficient in both training and prediction. In this work, we present a clinical-applicable MMD (>2) framework rFast-MMDNet, operating with raw projection data in non-recursive setup, for breast tissue differentiation.