High-quality semi-supervised anomaly detection with generative adversarial networks.

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

Abstract

PURPOSE: The visualization of an anomaly area is easier in anomaly detection methods that use generative models rather than classification models. However, achieving both anomaly detection accuracy and a clear visualization of anomalous areas is challenging. This study aimed to establish a method that combines both detection accuracy and clear visualization of anomalous areas using a generative adversarial network (GAN).

Authors

  • Yuki Sato
    Systems and Information Engineering Master's Program in Computer Science, University of Tsukuba, 1-1-1 Tenoudai, Tsukuba City, Ibaraki, 305-0821, Japan. s2220599@u.tsukuba.ac.jp.
  • Junya Sato
    Faculty of Medicine, Osaka University, 2-2 Yamadaoka, Suita, Osaka, 5650871, Japan.
  • Noriyuki Tomiyama
    Diagnostic and Interventional Radiology, Osaka University Graduate School of Medicine, Suita, Osaka, Japan.
  • Shoji Kido
    Graduate School of Medicine, Yamaguchi University, Tokiwadai 2-16-1, Ube, Yamaguchi, 755-8611, Japan.