期刊
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
卷 117, 期 48, 页码 30088-30095出版社
NATL ACAD SCIENCES
DOI: 10.1073/pnas.1907377117
关键词
instability; deep learning; AI; image reconstruction; inverse problems
资金
- European Union's Horizon 2020 research and innovation program under Marie Sklodowska-Curie Grant [655282]
- FCT (Fundacao para a Ciencia e a Tecnologia, I.P.) [CEECIND/01970/2017]
- Natural Sciences and Engineering Research Council of Canada (NSERC) [611675]
- Royal Society University Research Fellowship
- UK Engineering and Physical Sciences Research Council [EP/L003457/1]
- EPSRC [EP/L003457/1] Funding Source: UKRI
- Marie Curie Actions (MSCA) [655282] Funding Source: Marie Curie Actions (MSCA)
Deep learning, due to its unprecedented success in tasks such as image classification, has emerged as a new tool in image reconstruction with potential to change the field. In this paper, we demonstrate a crucial phenomenon: Deep learning typically yields unstable methods for image reconstruction. The instabilities usually occur in several forms: 1) Certain tiny, almost undetectable perturbations, both in the image and sampling domain, may result in severe artefacts in the reconstruction; 2) a small structural change, for example, a tumor, may not be captured in the reconstructed image; and 3) (a counterintuitive type of instability) more samples may yield poorer performance. Our stability test with algorithms and easy-to-use software detects the instability phenomena. The test is aimed at researchers, to test their networks for instabilities, and for government agencies, such as the Food and Drug Administration (FDA), to secure safe use of deep learning methods.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据