Investigation of Adversarial Robust Training for Establishing Interpretable CNN-based Numerical Observers

Sourya Sengupta, Craig K. Abbey, Kaiyan Li, Mark A. Anastasio

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The use of convolutional neural networks (CNNs) for establishing anthropomorphic numerical observers (ANOs) is being actively explored. In these data-driven approaches, CNNs are trained in a standard supervised way with human-labeled training data; hence, the anthropomorphic component of this procedure resides only in the training labels. However, it is well-known that such traditionally trained CNNs can rely on image features that are highly specific to the training distribution and may not align with features exploited by human perception. While being able to predict human observer performance under certain specified conditions, traditionally-Trained CNNs lack the interpretability and robustness that may be desired for an ANO. To address this, in this work we investigate the use of an adversarial robust training strategy for training CNN-based observers. As recently demonstrated in the computer vision literature, this training strategy can result in CNNs that exploit more human-interpretable features than would be employed by a standard CNN. Robustly trained CNNs are systematically investigated for performing a signal-known-exactly (SKE) and background-known-statistically (BKS) binary detection task. Additionally, a differential evolution-based optimization procedure is developed to establish robustly trained CNNs that achieve a specified performance, which may provide a new approach to establishing ANOs. 2022 SPIE.

Original languageEnglish (US)
Title of host publicationMedical Imaging 2022
Subtitle of host publicationImage Perception, Observer Performance, and Technology Assessment
EditorsClaudia R. Mello-Thoms, Claudia R. Mello-Thoms, Sian Taylor-Phillips
PublisherSPIE
ISBN (Electronic)9781510649453
DOIs
StatePublished - 2022
Externally publishedYes
EventMedical Imaging 2022: Image Perception, Observer Performance, and Technology Assessment - Virtual, Online
Duration: Mar 21 2022Mar 27 2022

Publication series

NameProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume12035
ISSN (Print)1605-7422

Conference

ConferenceMedical Imaging 2022: Image Perception, Observer Performance, and Technology Assessment
CityVirtual, Online
Period3/21/223/27/22

Keywords

  • Adversarial Robust Training
  • Classification Images
  • Numerical Observers

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Atomic and Molecular Physics, and Optics
  • Biomaterials
  • Radiology Nuclear Medicine and imaging

Fingerprint

Dive into the research topics of 'Investigation of Adversarial Robust Training for Establishing Interpretable CNN-based Numerical Observers'. Together they form a unique fingerprint.

Cite this