Enhancing MR image segmentation with realistic adversarial data augmentation
File(s)1-s2.0-S1361841522002304-main.pdf (2.94 MB)
Published version
Author(s)
Type
Journal Article
Abstract
The success of neural networks on medical image segmentation tasks typically relies on large labeled datasets for model training. However, acquiring and manually labeling a large medical image set is resource-intensive, expensive, and sometimes impractical due to data sharing and privacy issues. To address this challenge, we propose AdvChain, a generic adversarial data augmentation framework, aiming at improving both the diversity and effectiveness of training data for medical image segmentation tasks. AdvChain augments data with dynamic data augmentation, generating randomly chained photo-metric and geometric transformations to resemble realistic yet challenging imaging variations to expand training data. By jointly optimizing the data augmentation model and a segmentation network during training, challenging examples are generated to enhance network generalizability for the downstream task. The proposed adversarial data augmentation does not rely on generative networks and can be used as a plug-in module in general segmentation networks. It is computationally efficient and applicable for both low-shot supervised and semi-supervised learning. We analyze and evaluate the method on two MR image segmentation tasks: cardiac segmentation and prostate segmentation with limited labeled data. Results show that the proposed approach can alleviate the need for labeled data while improving model generalization ability, indicating its practical value in medical imaging applications.
Date Issued
2022-11
Date Acceptance
2022-08-19
ISSN
1361-8415
Publisher
Elsevier
Start Page
1
End Page
15
Journal / Book Title
Medical Image Analysis
Volume
82
Copyright Statement
© 2022 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Sponsor
Engineering and Physical Sciences Research Council
Engineering & Physical Science Research Council (E
Identifier
https://www.sciencedirect.com/science/article/pii/S1361841522002304?via%3Dihub
http://arxiv.org/abs/2108.03429v2
Grant Number
EP/W01842X/1
EP/W01842X/1
Subjects
Adversarial data augmentation
Adversarial training
Data augmentation
MR image segmentation
Model generalization
eess.IV
eess.IV
cs.CV
cs.LG
q-bio.QM
eess.IV
eess.IV
cs.CV
cs.LG
q-bio.QM
09 Engineering
11 Medical and Health Sciences
Nuclear Medicine & Medical Imaging
Notes
Under review. 23 pages
Publication Status
Published
OA Location
https://arxiv.org/pdf/2108.03429.pdf
Date Publish Online
2022-08-28