Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Oct 2, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

3.0K

Natural Images Allow Universal Adversarial Attacks on Medical Image Classification Using Deep Neural Networks with

Akinori Minagi1, Hokuto Hirano1, Kauzhiro Takemoto1

  • 1Department of Bioscience and Bioinformatics, Kyushu Institute of Technology, Iizuka 820-8502, Fukuoka, Japan.

Journal of Imaging
|February 24, 2022
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Adversarial control of synchronization in complex oscillator networks.

Chaos (Woodbury, N.Y.)·2025
Same author

Large-scale moral machine experiment on large language models.

PloS one·2025
Same author

The moral machine experiment on large language models.

Royal Society open science·2024
Same author

Universal adversarial attacks on deep neural networks for medical image classification.

BMC medical imaging·2021
Same author

Vulnerability of deep neural networks for detecting COVID-19 cases from chest X-ray images to universal adversarial attacks.

PloS one·2020
Same author

Revisiting the hypothesis of an energetic barrier to genome complexity between eukaryotes and prokaryotes.

Royal Society open science·2020
Same journal

Human-AI Interaction in Interventional Radiology: A Narrative Review of Current Applications, Challenges, and Future Directions.

Journal of imaging·2026
Same journal

Coronary Artery Anomalies and Anatomical Variants: Cross-Sectional Diagnostic Imaging and Clinical Background.

Journal of imaging·2026
Same journal

YoLeTooth: A Unified Framework for Joint Tooth Segmentation and Periapical Lesion Detection in Panoramic Radiographs.

Journal of imaging·2026
Same journal

Radiomics-Guided Multi-Sequence Learning for Pathological Complete Response Prediction from Breast MRI with Missing Auxiliary Sequences.

Journal of imaging·2026
Same journal

Cutaneous Thermography in Arthropathies: Quantitative Imaging, Machine Learning, and Clinical Translation.

Journal of imaging·2026
Same journal

Two-Stage Dynamic Synergistic Segmentation Method for Myocardial Pathology.

Journal of imaging·2026
See all related articles

Adversarial attacks on medical deep neural networks are possible using natural images, even without medical data. Transfer learning creates vulnerabilities, posing a significant security threat to AI-driven disease diagnosis.

Area of Science:

  • Artificial Intelligence
  • Medical Imaging
  • Cybersecurity

Background:

  • Deep neural networks (DNNs) with transfer learning are used for medical image classification.
  • Adversarial vulnerability of DNNs poses risks to clinical diagnosis.
  • Medical image datasets are typically unavailable for adversarial attacks due to privacy concerns.

Purpose of the Study:

  • To demonstrate adversarial attacks on medical DNNs using natural images.
  • To investigate the effectiveness of universal adversarial perturbations (UAPs) generated from natural images.
  • To assess the security implications of transfer learning in medical AI.

Main Methods:

  • Generating universal adversarial perturbations (UAPs) from natural images.
  • Applying UAPs to medical DNN models trained with transfer learning.
Keywords:
adversarial attacksdeep neural networksmedical imagingprivacysecuritytransfer learning

Related Experiment Videos

Last Updated: Oct 2, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

3.0K
  • Evaluating UAP performance against non-targeted and targeted attacks.
  • Comparing UAP performance with random controls and models trained from random initialization.

    • Main Results:

    • Adversarial attacks are feasible using natural images on medical DNNs with transfer learning.
    • UAPs generated from natural images are effective for both non-targeted and targeted attacks.
    • The performance of natural image UAPs significantly exceeds random controls.
    • Transfer learning introduces a security vulnerability, decreasing diagnostic reliability.

    Conclusions:

    • Transfer learning in medical DNNs creates a security loophole exploitable by natural image-based adversarial attacks.
    • This vulnerability threatens the reliability and safety of computer-aided disease diagnosis.
    • While random initialization reduces UAP effectiveness, it does not eliminate the vulnerability, highlighting a significant future security concern.