Robust convolutional neural networks against adversarial attacks on medical images. (December 2022)

• Record Type: Journal Article
• Authors: Shi, Xiaoshuang; Peng, Yifan; Chen, Qingyu; Keenan, Tiarnan; Thavikulwat, Alisa T.; Lee, Sungwon; Tang, Yuxing; Chew, Emily Y.; Summers, Ronald M.; Lu, Zhiyong
• Is Part Of: Pattern recognition. Volume 132(2022)
• Publication Date: 2022-12
• Subjects: CNNs -- Adversarial examples -- Sparsity denoising; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2022.108923
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: We quantify the scale of adversarial perturbations imperceptible to clinicians. Noise might cause CNNs' vulnerability to adversarial medical images. We propose sparsity denoising operators for boosting CNNs' robustness. Abstract: Convolutional neural networks (CNNs) have been widely applied to medical images. However, medical images are vulnerable to adversarial attacks by perturbations that are undetectable to human experts. This poses significant security risks and challenges to CNN-based applications in clinic practice. In this work, we quantify the scale of adversarial perturbation imperceptible to clinical practitioners and investigate the cause of the vulnerability in CNNs. Specifically, we discover that noise (i.e., irrelevant or corrupted discriminative information) in medical images might be a key contributor to performance deterioration of CNNs against adversarial perturbations, as noisy features are learned unconsciously by CNNs in feature representations and magnified by adversarial perturbations. In response, we propose a novel defense method by embedding sparsity denoising operators in CNNs for improved robustness. Tested with various state-of-the-art attacking methods on two distinct medical image modalities, we demonstrate that the proposed method can successfully defend against those unnoticeable adversarial attacks by retaining as much as over 90% of its original performance. We believe our findings are critical for improving and deploying CNN-based medical applications in real-world scenarios.