EUVIMEDEUVIMED entdecken
Crossref

GC 2 : Generalizable Continual Classification of Medical Images

Nourhan Bayasi, Ghassan Hamarneh, Rafeef Garbi

IEEE Transactions on Medical Imaging · 2024-11 · 43 (11) · 3767-3779

DOI: 10.1109/tmi.2024.3398533 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Medical Imaging, 2024-11. Verfasst von Nourhan Bayasi und 2 weiteren Autor:innen. Hauptthemen laut MEDLINE: Deep Learning, Algorithms. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 53 Literaturangaben. Gefördert von: Vanier Ph.D. Fellowship from the Natural Sciences and Engineering Research Council (NSERC) of Canada.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Deep learning models have achieved remarkable success in medical image classification. These models are typically trained once on the available annotated images and thus lack the ability of continually learning new tasks (i.e., new classes or data distributions) due to the problem of catastrophic forgetting. Recently, there has been more interest in designing continual learning methods to learn different tasks presented sequentially over time while preserving previously acquired knowledge. However, these methods focus mainly on preventing catastrophic forgetting and are tested under a closed-world assumption; i.e., assuming the test data is drawn from the same distribution as the training data. In this work, we advance the state-of-the-art in continual learning by proposing GC2 for medical image classification, which learns a sequence of tasks while simultaneously enhancing its out-of-distribution robustness. To alleviate forgetting, GC2 employs a gradual culpability-based network pruning to identify an optimal subnetwork for each task. To improve generalization, GC2 incorporates adversarial image augmentation and knowledge distillation approaches for learning generalized and robust representations for each subnetwork. Our extensive experiments on multiple benchmarks in a task-agnostic inference demonstrate that GC2 significantly outperforms baselines and other continual learning methods in reducing forgetting and enhancing generalization. Our code is publicly available at the following link: https://github.com/nourhanb/TMI2024-GC2.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningAlgorithmsHumansImage Interpretation, Computer-AssistedImage Processing, Computer-AssistedDatabases, Factual

Förderung

  • Vanier Ph.D. Fellowship from the Natural Sciences and Engineering Research Council (NSERC) of Canada

Zitieren

APA

Bayasi, N., Hamarneh, G., & Garbi, R. (2024). GC 2 : Generalizable Continual Classification of Medical Images. IEEE Transactions on Medical Imaging, 43(11), 3767-3779. https://doi.org/10.1109/tmi.2024.3398533

Vancouver

Bayasi N, Hamarneh G, Garbi R. GC 2 : Generalizable Continual Classification of Medical Images. IEEE Transactions on Medical Imaging. 2024;43(11):3767-3779. doi:10.1109/tmi.2024.3398533

BibTeX
@article{Bayasi2024,
  author = {Bayasi, Nourhan and Hamarneh, Ghassan and Garbi, Rafeef},
  title = {GC 2 : Generalizable Continual Classification of Medical Images},
  journal = {IEEE Transactions on Medical Imaging},
  year = {2024},
  volume = {43},
  number = {11},
  pages = {3767--3779},
  doi = {10.1109/tmi.2024.3398533}
}
Kostenlos starten →