KIAM Main page Web Library  •  Publication Searh   

Conference material: "Proceedings of the International Conference on Computer Graphics and Vision Graphicon"
Authors: Lockshin N.D., Khvostikov A.V., Krylov A.S.
Augmenting Histological Images with Adversarial Attacks
Neural networks have shown to be vulnerable against adversarial attacks - images with carefully crafted adversarial noise that is imperceptible to the human eye. In medical imaging tasks this can be a major threat for making predictions based on deep neural network solutions. In this paper we propose a pipeline for augmenting a small histological image dataset using State-of-the-Art data generation methods and demonstrate an increase in accuracy of a neural classifier trained on the augmented dataset when faced with adversarial images. When trained on the non-augmented dataset, the neural network achieves an accuracy of 55.24 on the test set with added adversarial noise, and an accuracy of 97.40 on the same test set when trained on the augmented dataset.
Adversarial Attacks, Deep Learning, Image Classification, Histology, Tissue Recognition
Publication language: english,  pages: 11 (p. 637-647)
English source text:
Export link to publication in format:   RIS    BibTeX
About authors:
  • Lockshin Nikita Djeffrievich,,  Lomonosov Moscow State University
  • Khvostikov Alexander Vladimirovich,,  Lomonosov Moscow State University
  • Krylov Andrey Serdjevich,,  Lomonosov Moscow State University