End-to-End Model-based Deep Learning for Dual-Energy Computed Tomography Material Decomposition

TL;DR

This paper introduces E2E-DEcomp, a novel deep learning method that directly converts DECT projection data into material images, integrating spectral knowledge and data-driven priors for improved material decomposition.

Contribution

The work presents an end-to-end deep learning framework for DECT material decomposition that does not require energy-based images during training, unlike previous methods.

Findings

E2E-DEcomp outperforms existing deep learning methods on spectral CT data.

The approach effectively incorporates spectral system knowledge into the training process.

The method simplifies training by only requiring sinogram and material images, not energy images.

Abstract

Dual energy X-ray Computed Tomography (DECT) enables to automatically decompose materials in clinical images without the manual segmentation using the dependency of the X-ray linear attenuation with energy. In this work we propose a deep learning procedure called End-to-End Material Decomposition (E2E-DEcomp) for quantitative material decomposition which directly convert the CT projection data into material images. The algorithm is based on incorporating the knowledge of the spectral model DECT system into the deep learning training loss and combining a data-learned prior in the material image domain. Furthermore, the training does not require any energy-based images in the dataset but rather only sinogram and material images. We show the effectiveness of the proposed direct E2E-DEcomp method on the AAPM spectral CT dataset (Sidky and Pan, 2023) compared with state of the art supervised deep learning networks.