Research reports

Years: 2024 2023 2022 2021 2020 2019 2018 2017 2016 2015 2014 2013 2012 2011 2010 2009 2008 2007 2006 2005 2004 2003 2002 2001 2000 1999 1998 1997 1996 1995 1994 1993 1992 1991

Localized adversarial artifacts for compressed sensing MRI

by R. Alaifari and G.S. Alberti and T. Gauksson

(Report number 2022-28)

Abstract
As interest in deep neural networks (DNNs) for image reconstruction tasks grows, their reliability has been called into question (Antun et al., 2020; Gottschling et al., 2020). However, recent work has shown that compared to total variation (TV) minimization, they show similar robustness to adversarial noise in terms of \(\ell^2\)-reconstruction error (Genzel et al., 2022). We consider a different notion of robustness, using the \(\ell^\infty\)-norm, and argue that localized reconstruction artifacts are a more relevant defect than the \(\ell^2\)-error. We create adversarial perturbations to undersampled MRI measurements which induce severe localized artifacts in the TV-regularized reconstruction. The same attack method is not as effective against DNN based reconstruction. Finally, we show that this phenomenon is inherent to reconstruction methods for which exact recovery can be guaranteed, as with compressed sensing reconstructions with \(\ell^1\)- or TV-minimization.

Keywords: Compressed sensing, magnetic resonance imaging, deep neural networks, total variation, adversarial examples.

@Techreport{AAG22_1016,
  author = {R. Alaifari and G.S. Alberti and T. Gauksson},
  title = {Localized adversarial artifacts for compressed sensing MRI},
  institution = {Seminar for Applied Mathematics, ETH Z{\"u}rich},
  number = {2022-28},
  address = {Switzerland},
  url = {https://www.sam.math.ethz.ch/sam_reports/reports_final/reports2022/2022-28.pdf },
  year = {2022}
}

Disclaimer
© Copyright for documents on this server remains with the authors. Copies of these documents made by electronic or mechanical means including information storage and retrieval systems, may only be employed for personal use. The administrators respectfully request that authors inform them when any paper is published to avoid copyright infringement. Note that unauthorised copying of copyright material is illegal and may lead to prosecution. Neither the administrators nor the Seminar for Applied Mathematics (SAM) accept any liability in this respect. The most recent version of a SAM report may differ in formatting and style from published journal version. Do reference the published version if possible (see SAM Publications).