跳到主要导航 跳到搜索 跳到主要内容

DBA: Downsampling-Based Adversarial Attack in Medical Image Analysis

  • Northwestern Polytechnical University Xian
  • CAS - Institute of Modern Physics

科研成果: 书/报告/会议事项章节 › 会议稿件 › 同行评审

1 引用 (Scopus)

摘要

The next-generation of artificial intelligence technology has contributed significantly to the development of medical intelligence. However, the widespread use of deep neural networks (DNNs) has also brought about serious security threats. In this paper, we present an adversarial attack approach for deep learning-based image segmentation models in the field of medical image analysis. In our solutions, we propose a novel adversarial attack method, which is designed to exploit the DNNs’ generic down-sampling operation to ensure the effectiveness, stealthiness, and transferability of the attack. We perform the attack on two State-Of-The-Art (SOTA) models, DDANet and CaraNet in a general medical image dataset Kvasir-SEG, and a comprehensive evaluation shows that our attack is effective stealthy, and transferrable.

源语言英语
主期刊名Third International Conference on Computer Vision and Pattern Analysis, ICCPA 2023
编辑Linlin Shen, Guoqiang Zhong
出版商SPIE
ISBN(电子版)9781510667563
DOI
出版状态已出版 - 2023
活动3rd International Conference on Computer Vision and Pattern Analysis, ICCPA 2023 - Hangzhou, 中国
期限: 7 4月 2023 → 9 4月 2023

丛书

姓名Proceedings of SPIE - The International Society for Optical Engineering
卷12754
ISSN(印刷版)0277-786X
ISSN(电子版)1996-756X

会议

会议3rd International Conference on Computer Vision and Pattern Analysis, ICCPA 2023
国家/地区中国
市Hangzhou
时期7/04/23 → 9/04/23

学术指纹

探究 'DBA: Downsampling-Based Adversarial Attack in Medical Image Analysis' 的科研主题。它们共同构成独一无二的学术指纹。

引用此