Infinite Brain MR Images: PGGAN-Based Data Augmentation for Tumor Detection

Changhee Han, Leonardo Rundo, Ryosuke Araki, Yujiro Furukawa, Giancarlo Mauri, Hideki Nakayama, Hideaki Hayashi

Research output: Chapter in Book/Report/Conference proceedingChapter

17 Citations (Scopus)

Abstract

Due to the lack of available annotated medical images, accurate computer-assisted diagnosis requires intensive data augmentation (DA) techniques, such as geometric/intensity transformations of original images; however, those transformed images intrinsically have a similar distribution to the original ones, leading to limited performance improvement. To fill the data lack in the real image distribution, we synthesize brain contrast-enhanced magnetic resonance (MR) images—realistic but completely different from the original ones—using generative adversarial networks (GANs). This study exploits progressive growing of GANs (PGGANs), a multistage generative training method, to generate original-sized 256 × 256 MR images for convolutional neural network-based brain tumor detection, which is challenging via conventional GANs; difficulties arise due to unstable GAN training with high resolution and a variety of tumors in size, location, shape, and contrast. Our preliminary results show that this novel PGGAN-based DA method can achieve a promising performance improvement, when combined with classical DA, in tumor detection and also in other medical imaging tasks.

Original languageEnglish
Title of host publicationSmart Innovation, Systems and Technologies
PublisherSpringer Science and Business Media Deutschland GmbH
Pages291-303
Number of pages13
DOIs
Publication statusPublished - 2020

Publication series

NameSmart Innovation, Systems and Technologies
Volume151
ISSN (Print)2190-3018
ISSN (Electronic)2190-3026

All Science Journal Classification (ASJC) codes

  • Decision Sciences(all)
  • Computer Science(all)

Fingerprint Dive into the research topics of 'Infinite Brain MR Images: PGGAN-Based Data Augmentation for Tumor Detection'. Together they form a unique fingerprint.

Cite this