X-ray computed tomography (CT) is widely used in clinical practice. The involved ionizing\nX-ray radiation, however, could increase cancer risk. Hence, the reduction of the radiation dose\nhas been an important topic in recent years. Few-view CT image reconstruction is one of the main\nways to minimize radiation dose and potentially allow a stationary CT architecture. In this paper,\nwe propose a deep encoder-decoder adversarial reconstruction (DEAR) network for 3D CT image\nreconstruction from few-view data. Since the artifacts caused by few-view reconstruction appear in\n3D instead of 2D geometry, a 3D deep network has a great potential for improving the image quality\nin a data driven fashion. More specifically, our proposed DEAR-3D network aims at reconstructing\n3D volume directly from clinical 3D spiral cone-beam image data. DEAR is validated on a publicly\navailable abdominal CT dataset prepared and authorized by Mayo Clinic. Compared with other\n2D deep learning methods, the proposed DEAR-3D network can utilize 3D information to produce\npromising reconstruction results.
Loading....