Scan: Structure correcting adversarial network for organ segmentation in chest x-rays

Wei Dai, Nanqing Dong, Zeya Wang, Xiaodan Liang, Hao Zhang, Eric P. Xing

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

106 Scopus citations

Abstract

Chest X-ray (CXR) is one of the most commonly prescribed medical imaging procedures, often with over 2–10x more scans than other imaging modalities. These voluminous CXR scans place significant workloads on radiologists and medical practitioners. Organ segmentation is a key step towards effective computer-aided detection on CXR. In this work, we propose Structure Correcting Adversarial Network (SCAN) to segment lung fields and the heart in CXR images. SCAN incorporates a critic network to impose on the convolutional segmentation network the structural regularities inherent in human physiology. Specifically, the critic network learns the higher order structures in the masks in order to discriminate between the ground truth organ annotations from the masks synthesized by the segmentation network. Through an adversarial process, the critic network guides the segmentation network to achieve more realistic segmentation that mimics the ground truth. Extensive evaluation shows that our method produces highly accurate and realistic segmentation. Using only very limited training data available, our model reaches human-level performance without relying on any pre-trained model. Our method surpasses the current state-of-the-art and generalizes well to CXR images from different patient populations and disease profiles.

Original languageEnglish
Title of host publicationDeep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support - 4th International Workshop, DLMIA 2018 and 8th International Workshop, ML-CDS 2018 Held in Conjunction with MICCAI 2018
EditorsLena Maier-Hein, Tanveer Syeda-Mahmood, Zeike Taylor, Zhi Lu, Danail Stoyanov, Anant Madabhushi, João Manuel R.S. Tavares, Jacinto C. Nascimento, Mehdi Moradi, Anne Martel, Joao Paulo Papa, Sailesh Conjeti, Vasileios Belagiannis, Hayit Greenspan, Gustavo Carneiro, Andrew Bradley
Pages263-273
Number of pages11
DOIs
StatePublished - 2018
Event4th International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2018 and 8th International Workshop on Multimodal Learning for Clinical Decision Support, ML-CDS 2018 Held in Conjunction with MICCAI 2018 - Granada, Spain
Duration: Sep 20 2018Sep 20 2018

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11045 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference4th International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2018 and 8th International Workshop on Multimodal Learning for Clinical Decision Support, ML-CDS 2018 Held in Conjunction with MICCAI 2018
Country/TerritorySpain
CityGranada
Period9/20/189/20/18

Bibliographical note

Publisher Copyright:
© Springer Nature Switzerland AG 2018.

Keywords

  • Adversarial learning
  • Chest X-ray
  • Deep neural networks
  • Medical image segmentation

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Scan: Structure correcting adversarial network for organ segmentation in chest x-rays'. Together they form a unique fingerprint.

Cite this