MI$$^2$$ GAN: Generative Adversarial Network for Medical Image Domain Adaptation Using Mutual Information Constraint

Xinpeng Xie, Jiawei Chen, Yuexiang Li, Linlin Shen, Kai Ma, Yefeng Zheng

Research output: Chapter in Book/Conference proceedingConference contributionpeer-review

5 Citations (Scopus)

Abstract

Domain shift between medical images from multicentres is still an open question for the community, which degrades the generalization performance of deep learning models. Generative adversarial network (GAN), which synthesize plausible images, is one of the potential solutions to address the problem. However, the existing GAN-based approaches are prone to fail at preserving image-objects in image-to-image (I2I) translation, which reduces their practicality on domain adaptation tasks. In this paper, we propose a novel GAN (namely MIGAN) to maintain image-contents during cross-domain I2I translation. Particularly, we disentangle the content features from domain information for both the source and translated images, and then maximize the mutual information between the disentangled content features to preserve the image-objects. The proposed MIGAN is evaluated on two tasks—polyp segmentation using colonoscopic images and the segmentation of optic disc and cup in fundus images. The experimental results demonstrate that the proposed MIGAN can not only generate elegant translated images, but also significantly improve the generalization performance of widely used deep learning networks (e.g., U-Net).

Original languageEnglish
Title of host publicationMedical Image Computing and Computer Assisted Intervention – MICCAI 2020 - 23rd International Conference, Proceedings
EditorsAnne L. Martel, Purang Abolmaesumi, Danail Stoyanov, Diana Mateus, Maria A. Zuluaga, S. Kevin Zhou, Daniel Racoceanu, Leo Joskowicz
PublisherSpringer Science and Business Media Deutschland GmbH
Pages516-525
Number of pages10
ISBN (Print)9783030597122
DOIs
Publication statusPublished - 2020
Externally publishedYes
Event23rd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2020 - Lima, Peru
Duration: 4 Oct 20208 Oct 2020

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume12262 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference23rd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2020
Country/TerritoryPeru
CityLima
Period4/10/208/10/20

Keywords

  • Domain adaptation
  • Mutual information

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science (all)

Fingerprint

Dive into the research topics of 'MI$$^2$$ GAN: Generative Adversarial Network for Medical Image Domain Adaptation Using Mutual Information Constraint'. Together they form a unique fingerprint.

Cite this