论文标题
对脂肪肝疾病分类的深度学习模型的对抗性攻击通过修改超声图像重建方法
Adversarial attacks on deep learning models for fatty liver disease classification by modification of ultrasound image reconstruction method
论文作者
论文摘要
卷积神经网络（CNN）在医学图像分析任务中取得了显着成功。在超声（US）成像中，CNN已应用于对象分类，图像重建和组织表征。但是，CNN可能容易受到对抗攻击的攻击，即使应用于输入数据的小扰动也可能会显着影响模型性能并导致错误的输出。在这项工作中，我们设计了一种新颖的对抗性攻击，特定于超声（US）成像。美国图像是根据射频信号重建的。由于美国图像的外观取决于应用的图像重建方法，因此我们通过扰动US B模式图像重建方法来探索欺骗深度学习模型的可能性。我们应用零订单优化，以找到与衰减补偿和振幅压缩有关的图像重建参数的少量扰动，这可能导致错误的输出。我们使用用于脂肪肝病诊断的深度学习模型来说明我们的方法，拟议的对抗性攻击达到了48％的成功率。
Convolutional neural networks (CNNs) have achieved remarkable success in medical image analysis tasks. In ultrasound (US) imaging, CNNs have been applied to object classification, image reconstruction and tissue characterization. However, CNNs can be vulnerable to adversarial attacks, even small perturbations applied to input data may significantly affect model performance and result in wrong output. In this work, we devise a novel adversarial attack, specific to ultrasound (US) imaging. US images are reconstructed based on radio-frequency signals. Since the appearance of US images depends on the applied image reconstruction method, we explore the possibility of fooling deep learning model by perturbing US B-mode image reconstruction method. We apply zeroth order optimization to find small perturbations of image reconstruction parameters, related to attenuation compensation and amplitude compression, which can result in wrong output. We illustrate our approach using a deep learning model developed for fatty liver disease diagnosis, where the proposed adversarial attack achieved success rate of 48%.
