The Security of Deep Learning Defences for Medical Imaging
This addresses a critical security problem for healthcare providers using deep learning in medical imaging, highlighting that current defences are insufficient against adaptive attackers.
The paper tackles the vulnerability of deep learning defences in medical imaging to informed adversarial attacks, showing that an attacker can evade five state-of-the-art defences and successfully fool the model, rendering them ineffective.
Deep learning has shown great promise in the domain of medical image analysis. Medical professionals and healthcare providers have been adopting the technology to speed up and enhance their work. These systems use deep neural networks (DNN) which are vulnerable to adversarial samples; images with imperceivable changes that can alter the model's prediction. Researchers have proposed defences which either make a DNN more robust or detect the adversarial samples before they do harm. However, none of these works consider an informed attacker which can adapt to the defence mechanism. We show that an informed attacker can evade five of the current state of the art defences while successfully fooling the victim's deep learning model, rendering these defences useless. We then suggest better alternatives for securing healthcare DNNs from such attacks: (1) harden the system's security and (2) use digital signatures.