English

Image-to-Graph Convolutional Network for Deformable Shape Reconstruction from a Single Projection Image

Image and Video Processing 2021-11-02 v2 Computer Vision and Pattern Recognition

Abstract

Shape reconstruction of deformable organs from two-dimensional X-ray images is a key technology for image-guided intervention. In this paper, we propose an image-to-graph convolutional network (IGCN) for deformable shape reconstruction from a single-viewpoint projection image. The IGCN learns relationship between shape/deformation variability and the deep image features based on a deformation mapping scheme. In experiments targeted to the respiratory motion of abdominal organs, we confirmed the proposed framework with a regularized loss function can reconstruct liver shapes from a single digitally reconstructed radiograph with a mean distance error of 3.6mm.

Keywords

Cite

@article{arxiv.2108.12533,
  title  = {Image-to-Graph Convolutional Network for Deformable Shape Reconstruction from a Single Projection Image},
  author = {M. Nakao and F. Tong and M. Nakamura and T. Matsuda},
  journal= {arXiv preprint arXiv:2108.12533},
  year   = {2021}
}

Comments

This paper will be appeared in MICCAI 2021

R2 v1 2026-06-24T05:29:10.985Z