Unsupervised colonoscopic depth estimation by domain translations with a Lambertian-reflection keeping auxiliary task.

Journal: International journal of computer assisted radiology and surgery
Published Date: May 17, 2021

Abstract

PURPOSE: A three-dimensional (3D) structure extraction technique viewed from a two-dimensional image is essential for the development of a computer-aided diagnosis (CAD) system for colonoscopy. However, a straightforward application of existing depth-estimation methods to colonoscopic images is impossible or inappropriate due to several limitations of colonoscopes. In particular, the absence of ground-truth depth for colonoscopic images hinders the application of supervised machine learning methods. To circumvent these difficulties, we developed an unsupervised and accurate depth-estimation method.

Authors

  • Hayato Itoh
    Graduate School of Informatics, Nagoya University, Furo-cho, Chikusa-ku, Nagoya, 464-8601, Japan.
  • Masahiro Oda
    Graduate School of Informatics, Nagoya University, Furo-cho, Chikusa-ku, Nagoya, 464-8601, Japan.
  • Yuichi Mori
    Clinical Effectiveness Research Group, Institute of Health and Society, University of Oslo, and Department of Transplantation Medicine Oslo University Hospital, Oslo, Norway; Digestive Disease Center, Showa University Northern Yokohama Hospital, Yokohama, Japan. Electronic address: ibusiginjp@gmail.com.
  • Masashi Misawa
    Digestive Disease Center, Showa University Northern Yokohama Hospital, Tsuzuki-ku, Yokohama, 224-8503, Japan.
  • Shin-Ei Kudo
    Digestive Disease Center, Showa University Northern Yokohama Hospital, Tsuzuki-ku, Yokohama, 224-8503, Japan.
  • Kenichiro Imai
    Division of Endoscopy, Shizuoka Cancer Center, Shizuoka, Japan.
  • Sayo Ito
    Division of Endoscopy, Shizuoka Cancer Center, 1007 Shimonagakubo, Nagaizumi-cho, Sunto-gun, Shizuoka, 411-8777, Japan.
  • Kinichi Hotta
    Shizuoka Cancer Center, Shizuoka, Japan (K.H.).
  • Hirotsugu Takabatake
    Department of Respiratory Medicine, Sapporo-Minami-Sanjo Hospital, Nishi-6-chome, Minami-3-jo, Chuo-ku, Sapporo, Hokkaido, 060-0063, Japan.
  • Masaki Mori
    Department of Gastroenterological Surgery, Osaka University Graduate School of Medicine, Suita City, Osaka, Japan.
  • Hiroshi Natori
    Department of Respiratory Medicine, Keiwakai Nishioka Hospital, 1-52, 4-jo 4-chome, Nishioka, Toyohira-ku, Sapporo, Hokkaido, 062-0034, Japan.
  • Kensaku Mori
    Graduate School of Informatics, Nagoya University, Furo-cho, Chikusa-ku, Nagoya, 464-8601, Japan.

Keywords

External Resources

Popular Topics
Recent Journals