Understanding adversarial attacks on deep learning based medical image analysis systems

•Medical image DNNs are easier to be attacked than natural non-medical image DNNs.•Complex biological textures of medical images may lead to more vulnerable regions.•State-of-the-art deep networks can be overparameterized for medical imaging tasks.•Medical image adversarial attacks can also be easil...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Pattern recognition Jg. 110; S. 107332
Hauptverfasser:	Ma, Xingjun, Niu, Yuhao, Gu, Lin, Wang, Yisen, Zhao, Yitian, Bailey, James, Lu, Feng
Format:	Journal Article
Sprache:	Englisch
Veröffentlicht:	Elsevier Ltd 01.02.2021
Schlagworte:	Adversarial attack Adversarial example detection Deep learning Medical image analysis Deep learning Adversarial attack Adversarial example detection Medical image analysis
ISSN:	0031-3203, 1873-5142
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!