Adversarial deep structured nets for mass segmentation from mammograms

Wentao Zhu, Xiang Xiang, Trac D. Tran, Gregory D. Hager, Xiaohui Xie

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Mass segmentation provides effective morphological features which are important for mass diagnosis. In this work, we propose a novel end-to-end network for mammographic mass segmentation which employs a fully convolutional network (FCN) to model a potential function, followed by a conditional random field (CRF) to perform structured learning. Because the mass distribution varies greatly with pixel position, the FCN is combined with a position priori. Further, we employ adversarial training to eliminate over-fitting due to the small sizes of mammogram datasets. Multi-scale FCN is employed to improve the segmentation performance. Experimental results on two public datasets, IN breast and DDSM-BCRP, demonstrate that our end-to-end network achieves better performance than state-of-the-art approaches.1

Original languageEnglish (US)
Title of host publication2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018
PublisherIEEE Computer Society
Pages847-850
Number of pages4
ISBN (Electronic)9781538636367
DOIs
StatePublished - May 23 2018
Event15th IEEE International Symposium on Biomedical Imaging, ISBI 2018 - Washington, United States
Duration: Apr 4 2018Apr 7 2018

Publication series

NameProceedings - International Symposium on Biomedical Imaging
Volume2018-April
ISSN (Print)1945-7928
ISSN (Electronic)1945-8452

Other

Other15th IEEE International Symposium on Biomedical Imaging, ISBI 2018
CountryUnited States
CityWashington
Period4/4/184/7/18

Keywords

  • Adversarial deep structured networks
  • Adversarial fully convolutional networks
  • Segmentation

ASJC Scopus subject areas

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging

Fingerprint Dive into the research topics of 'Adversarial deep structured nets for mass segmentation from mammograms'. Together they form a unique fingerprint.

Cite this