A4 Article in conference proceedings
One-Pixel Attack Deceives Computer-Assisted Diagnosis of Cancer (2021)
Korpihalkola, J., Sipola, T., Puuska, S., & Kokkonen, T. (2021). One-Pixel Attack Deceives Computer-Assisted Diagnosis of Cancer. In SPML 2021 : 4th International Conference on Signal Processing and Machine Learning (pp. 100-106). ACM. https://doi.org/10.1145/3483207.3483224
JYU authors or editors
Publication details
All authors or editors: Korpihalkola, Joni; Sipola, Tuomo; Puuska, Samir; Kokkonen, Tero
Parent publication: SPML 2021 : 4th International Conference on Signal Processing and Machine Learning
Conference:
- International Conference on Signal Processing and Machine Learning
Place and date of conference: Beijing, China, 18.-20.8.2021
eISBN: 978-1-4503-9017-0
Publication year: 2021
Publication date: 18/08/2021
Pages range: 100-106
Number of pages in the book: 185
Publisher: ACM
Place of Publication: New York
Publication country: United States
Publication language: English
DOI: https://doi.org/10.1145/3483207.3483224
Publication open access: Not open
Publication channel open access:
Publication is parallel published (JYX): https://jyx.jyu.fi/handle/123456789/78557
Publication is parallel published: https://arxiv.org/abs/2012.00517
Abstract
Computer vision and machine learning can be used to automate various tasks in cancer diagnostic and detection. If an attacker can manipulate the automated processing, the results can be devastating and in the worst case lead to wrong diagnosis and treatment. In this research, the goal is to demonstrate the use of one-pixel attacks in a real-life scenario with a real pathology dataset, TUPAC16, which consists of digitized whole-slide images. We attack against the IBM CODAIT's MAX breast cancer detector using adversarial images. These adversarial examples are found using differential evolution to perform the one-pixel modification to the images in the dataset. The results indicate that a minor one-pixel modification of a whole slide image under analysis can affect the diagnosis by reversing the automatic diagnosis result. The attack poses a threat from the cyber security perspective: the one-pixel method can be used as an attack vector by a motivated attacker.
Keywords: computer vision; machine learning; diagnostics; cancerous diseases; cyber security; cyber attacks
Contributing organizations
Ministry reporting: Yes
Reporting Year: 2021
JUFO rating: 1