The effects of different levels of realism on the training of CNNs with only synthetic images for the semantic segmentation of robotic instruments in a head phantom.

Journal: International journal of computer assisted radiology and surgery
Published Date:

Abstract

PURPOSE: The manual generation of training data for the semantic segmentation of medical images using deep neural networks is a time-consuming and error-prone task. In this paper, we investigate the effect of different levels of realism on the training of deep neural networks for semantic segmentation of robotic instruments. An interactive virtual-reality environment was developed to generate synthetic images for robot-aided endoscopic surgery. In contrast with earlier works, we use physically based rendering for increased realism.

Authors

  • Saul Alexis Heredia Perez
    Department of Mechanical Engineering, The University of Tokyo, Tokyo, Japan.
  • Murilo Marques Marinho
    Department of Mechanical Engineering, The University of Tokyo, Tokyo, Japan. murilo@nml.t.u-tokyo.ac.jp.
  • Kanako Harada
    Department of Mechanical Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-8656, Japan.
  • Mamoru Mitsuishi
    Department of Mechanical Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-8656, Japan.