Double-mix pseudo-label framework: enhancing semi-supervised segmentation on category-imbalanced CT volumes.

Journal: International journal of computer assisted radiology and surgery
PMID:

Abstract

PURPOSE: Deep-learning-based supervised CT segmentation relies on fully and densely labeled data, the labeling process of which is time-consuming. In this study, our proposed method aims to improve segmentation performance on CT volumes with limited annotated data by considering category-wise difficulties and distribution.

Authors

  • Luyang Zhang
    Graduate School of Informatics, Nagoya University, Furo-cho, Chikusa-ku, 464-8601, Nagoya, Aichi, Japan. lzhang@mori.m.is.nagoya-u.ac.jp.
  • Yuichiro Hayashi
    Nagoya University, Furo-cho, Chikusa-ku, Nagoya, Japan.
  • Masahiro Oda
    Graduate School of Informatics, Nagoya University, Furo-cho, Chikusa-ku, Nagoya, 464-8601, Japan.
  • Kensaku Mori
    Graduate School of Informatics, Nagoya University, Furo-cho, Chikusa-ku, Nagoya, 464-8601, Japan.