Image-to-Volume Deformable Registration by Learning Displacement Vector Fields

Abstract

2-D/3-D image registration is a problem that solves the deformation and alignment of a pretreatment 3-D volume to a 2-D projection image, which is available for treatment support and biomedical analysis. 2-D/3-D image registration for abdominal organs is a complicated task because the abdominal organs deform significantly and their contours are not detected in 2-D X-ray images. In this study, we propose a supervised deep learning framework that achieves 2-D/3-D deformable image registration between the 3-D volume and a single-viewpoint 2-D projection image. The proposed method uses latent image features of the 2-D projection images to learn a transformation from the input image, which is a concatenation of the 2-D projection images and the 3-D volume, to a dense displacement vector field (DVF) that represents nonlinear and local organ displacements. The target DVFs are generated by registration between 3-D volumes, and the registration error with the estimated DVF is introduced as a loss function during training. We register 3D-computed tomography (CT) volumes to the digitally reconstructed radiographs generated from abdominal 4D-CT volumes of 35 cases. The experimental results show that the proposed method can reconstruct 3D-CT with a mean voxel-to-voxel error of 29.4 Hounsfield unit and a dice similarity coefficient of 89.2 % on average for the body, liver, stomach, duodenum, and kidney regions, which is a clinically acceptable accuracy. In addition, the average computation time for the registration process by the proposed framework is 0.181 s, demonstrating real-time registration performance.