A Multimodal Adversarial Database: Towards A Comprehensive Assessment of Adversarial Attacks and Defenses on Medical Images

Deep learning models have been widely applied in many fields, including medical image analysis and computer-aided disease diagnosis. However, these models are easily fooled by adversarial attacks from some created adversarial examples which are hardly distinguished by humans. In this paper, we implement a comprehensive assessment of six popular adversarial attacks on four multimodal medical image datasets using two main deep learning-based target models. Moreover, in order to evaluate the capability of defense, two new defense methods are leveraged to cope with medical adversarial attacks. More importantly, we also build and release a big multimodal medical adversarial database (including four medical adversarial datasets) with 712,596 examples to facilitate future research of adversarial attacks and defenses in the multimodal medical image field. Extensive experiments indicate that all-sided adversarial attacks like BIM are still scarce under different evaluation metrics and defenses are not universally successful.