Adversarial attacks on deep learning models for fatty liver disease classification by modification of ultrasound image reconstruction method

by   Michał Byra, et al.

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


page 1

page 2

page 3


Impact of ultrasound image reconstruction method on breast lesion classification with neural transfer learning

Deep learning algorithms, especially convolutional neural networks, have...

Realistic Ultrasound Image Synthesis for Improved Classification of Liver Disease

With the success of deep learning-based methods applied in medical image...

Data Stealing Attack on Medical Images: Is it Safe to Export Networks from Data Lakes?

In privacy-preserving machine learning, it is common that the owner of t...

A Study for Universal Adversarial Attacks on Texture Recognition

Given the outstanding progress that convolutional neural networks (CNNs)...

On instabilities of deep learning in image reconstruction - Does AI come at a cost?

Deep learning, due to its unprecedented success in tasks such as image c...

Adversarial Robustness of MR Image Reconstruction under Realistic Perturbations

Deep Learning (DL) methods have shown promising results for solving ill-...

A comprehensive benchmark analysis for sand dust image reconstruction

Numerous sand dust image enhancement algorithms have been proposed in re...