Repository logo

Infoscience

  • English
  • French
Log In
Logo EPFL, École polytechnique fédérale de Lausanne

Infoscience

  • English
  • French
Log In
  1. Home
  2. Academic and Research Output
  3. Conferences, Workshops, Symposiums, and Seminars
  4. AMAE: Adaptation of Pre-trained Masked Autoencoder for Dual-Distribution Anomaly Detection in Chest X-Rays
 
Loading...
Thumbnail Image
conference paper

AMAE: Adaptation of Pre-trained Masked Autoencoder for Dual-Distribution Anomaly Detection in Chest X-Rays

Bozorgtabar, Behzad  
•
Mahapatra, Dwarikanath
•
Thiran, Jean-Philippe  
Madabhushi, A
•
Greenspan, H
Show more
January 1, 2023
Medical Image Computing And Computer Assisted Intervention, Miccai 2023, Pt I
26th International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI)

Unsupervised anomaly detection in medical images such as chest radiographs is stepping into the spotlight as it mitigates the scarcity of the labor-intensive and costly expert annotation of anomaly data. However, nearly all existing methods are formulated as a one-class classification trained only on representations from the normal class and discard a potentially significant portion of the unlabeled data. This paper focuses on a more practical setting, dual distribution anomaly detection for chest X-rays, using the entire training data, including both normal and unlabeled images. Inspired by a modern self-supervised vision transformer model trained using partial image inputs to reconstruct missing image regions- we propose AMAE, a two-stage algorithm for adaptation of the pre-trained masked autoencoder (MAE). Starting from MAE initialization, AMAE first creates synthetic anomalies from only normal training images and trains a lightweight classifier on frozen transformer features. Subsequently, we propose an adaptation strategy to leverage unlabeled images containing anomalies. The adaptation scheme is accomplished by assigning pseudo-labels to unlabeled images and using two separate MAE based modules to model the normative and anomalous distributions of pseudo-labeled images. The effectiveness of the proposed adaptation strategy is evaluated with different anomaly ratios in an unlabeled training set. AMAE leads to consistent performance gains over competing self-supervised and dual distribution anomaly detection methods, setting the new state-of-the-art on three public chest X-ray benchmarks - RSNA, NIH-CXR, and VinDr-CXR.

  • Details
  • Metrics
Type
conference paper
DOI
10.1007/978-3-031-43907-0_19
Web of Science ID

WOS:001109628700019

Author(s)
Bozorgtabar, Behzad  
•
Mahapatra, Dwarikanath
•
Thiran, Jean-Philippe  
Editors
Madabhushi, A
•
Greenspan, H
•
Mousavi, P
•
Salcudean, S
•
Duncan, J
•
Syeda-Mahmood, T
•
Taylor, R
Date Issued

2023-01-01

Publisher

Springer International Publishing Ag

Publisher place

Cham

Published in
Medical Image Computing And Computer Assisted Intervention, Miccai 2023, Pt I
ISBN of the book

978-3-031-43906-3

978-3-031-43907-0

Volume

14220

Start page

195

End page

205

Subjects

Technology

•

Life Sciences & Biomedicine

•

Anomaly Detection

•

Chest X-Ray

•

Masked Autoencoder

Peer reviewed

REVIEWED

Written at

EPFL

Event nameEvent placeEvent date
26th International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI)

Vancouver, CANADA

OCT 08-12, 2023

Available on Infoscience
March 18, 2024
Use this identifier to reference this record
https://infoscience.epfl.ch/handle/20.500.14299/206283
Logo EPFL, École polytechnique fédérale de Lausanne
  • Contact
  • infoscience@epfl.ch

  • Follow us on Facebook
  • Follow us on Instagram
  • Follow us on LinkedIn
  • Follow us on X
  • Follow us on Youtube
AccessibilityLegal noticePrivacy policyCookie settingsEnd User AgreementGet helpFeedback

Infoscience is a service managed and provided by the Library and IT Services of EPFL. © EPFL, tous droits réservés