EUVIMEDEUVIMED entdecken
Crossref

Encoder-Decoder Contrast for Unsupervised Anomaly Detection in Medical Images

Jia Guo, Shuai Lu, Lize Jia, Weihang Zhang, Huiqi Li

IEEE Transactions on Medical Imaging · 2024-03 · 43 (3) · 1102-1112

DOI: 10.1109/tmi.2023.3327720 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Medical Imaging, 2024-03. Verfasst von Jia Guo und 4 weiteren Autor:innen. Hauptthemen laut MEDLINE: Neuroimaging, Tomography, Optical Coherence. Die Arbeit wurde bisher 44-mal zitiert (Crossref) und nennt 52 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Beijing Natural Science Foundation, Beijing Institute of Technology Research Fund Program for Young Scholars.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Unsupervised anomaly detection (UAD) aims to recognize anomalous images based on the training set that contains only normal images. In medical image analysis, UAD benefits from leveraging the easily obtained normal (healthy) images, avoiding the costly collecting and labeling of anomalous (unhealthy) images. Most advanced UAD methods rely on frozen encoder networks pre-trained using ImageNet for extracting feature representations. However, the features extracted from the frozen encoders that are borrowed from natural image domains coincide little with the features required in the target medical image domain. Moreover, optimizing encoders usually causes pattern collapse in UAD. In this paper, we propose a novel UAD method, namely Encoder-Decoder Contrast (EDC), which optimizes the entire network to reduce biases towards pre-trained image domain and orient the network in the target medical domain. We start from feature reconstruction approach that detects anomalies from reconstruction errors. Essentially, a contrastive learning paradigm is introduced to tackle the problem of pattern collapsing while optimizing the encoder and the reconstruction decoder simultaneously. In addition, to prevent instability and further improve performances, we propose to bring globality into the contrastive objective function. Extensive experiments are conducted across four medical image modalities including optical coherence tomography, color fundus image, brain MRI, and skin lesion image, where our method outperforms all current state-of-the-art UAD methods. Code is available at: https://github.com/guojiajeremy/EDC.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

NeuroimagingTomography, Optical CoherenceFundus OculiImage Processing, Computer-Assisted

Förderung

  • National Natural Science Foundation of China
  • Beijing Natural Science Foundation
  • Beijing Institute of Technology Research Fund Program for Young Scholars

Zitieren

APA

Guo, J., Lu, S., Jia, L., Zhang, W., & Li, H. (2024). Encoder-Decoder Contrast for Unsupervised Anomaly Detection in Medical Images. IEEE Transactions on Medical Imaging, 43(3), 1102-1112. https://doi.org/10.1109/tmi.2023.3327720

Vancouver

Guo J, Lu S, Jia L, Zhang W, Li H. Encoder-Decoder Contrast for Unsupervised Anomaly Detection in Medical Images. IEEE Transactions on Medical Imaging. 2024;43(3):1102-1112. doi:10.1109/tmi.2023.3327720

BibTeX
@article{Guo2024,
  author = {Guo, Jia and Lu, Shuai and Jia, Lize and Zhang, Weihang and Li, Huiqi},
  title = {Encoder-Decoder Contrast for Unsupervised Anomaly Detection in Medical Images},
  journal = {IEEE Transactions on Medical Imaging},
  year = {2024},
  volume = {43},
  number = {3},
  pages = {1102--1112},
  doi = {10.1109/tmi.2023.3327720}
}
Kostenlos starten →