Structure and Illumination Constrained GAN for Medical Image Enhancement

Yuhui Ma, Jiang Liu, YONGHUAI LIU, Huazhu Fu, Yan Hu, Jun Cheng, Hong Qi, Yufei Wu, Jong Zhang, Yitian Zhao*

*Corresponding author for this work

Research output: Contribution to journalArticle (journal)peer-review

124 Citations (Scopus)
1632 Downloads (Pure)

Abstract

The development of medical imaging techniques
has greatly supported clinical decision making.
However, poor imaging quality, such as non-uniform illumination
or imbalanced intensity, brings challenges for
automated screening, analysis and diagnosis of diseases.
Previously, bi-directional GANs (e.g., CycleGAN), have been
proposed to improve the quality of input images without
the requirement of paired images. However, these methods
focus on global appearance, without imposing constraints
on structure or illumination, which are essential features
for medical image interpretation. In this paper, we propose
a novel and versatile bi-directional GAN, named Structure
and illumination constrained GAN (StillGAN), for medical
image quality enhancement. Our StillGAN treats low- and
high-quality images as two distinct domains, and introduces
local structure and illumination constraints for learning
both overall characteristics and local details. Extensive
experiments on three medical image datasets (e.g., corneal
confocal microscopy, retinal color fundus and endoscopy
images) demonstrate that our method performs better than
both conventional methods and other deep learning-based
methods. In addition, we have investigated the impact of
the proposed method on different medical image analysis
and clinical tasks such as nerve segmentation, tortuosity
grading, fovea localization and disease classification.
Original languageEnglish
Pages (from-to)3955-3967
Number of pages13
JournalIEEE Transactions on Medical Imaging
Volume40
Issue number12
Early online date2 Aug 2021
DOIs
Publication statusPublished - 1 Dec 2021

Keywords

  • Bi-directional GAN
  • Illumination regularization
  • structure loss
  • medical image enhancement
  • illumination regularization

Research Groups

  • Visual Computing Lab

Fingerprint

Dive into the research topics of 'Structure and Illumination Constrained GAN for Medical Image Enhancement'. Together they form a unique fingerprint.

Cite this