Abstract
Clinically Relevant Vulnerabilities of Deep Machine Learning Systems for Skin Cancer Diagnosis
Highlights
We report two, to our knowledge, previously unreported classes of relatively simple adversarial attacks that could arise inadvertently in a clinical setting: (i) alterations in color balance and (ii) alterations in rotation and/or translation of the input image that lead to misclassification of melanoma as a benign nevus
(Figure 1hem; Supplementary Materials and Methods), which modifies only three pixels within the input image leaving all others unchanged. We found that this method led to successful adversarial attacks; in comparison with the fast gradient sign method, there was a lower success rate, and attacks were only successful when the initial confidence of the network in predicting melanoma was lower for a particular image (Figure 1n)
To explore whether alterations in the overall image color balance could influence the accuracy of skin cancer lesion classification by a Convolutional neural networks (CNNs), we employed a differential evolution algorithm to search for subtle perturbations of global color balance that could lead to misclassification (Figure 2a)
Summary
We report two, to our knowledge, previously unreported classes of relatively simple adversarial attacks that could arise inadvertently in a clinical setting: (i) alterations in color balance and (ii) alterations in rotation and/or translation of the input image that lead to misclassification of melanoma as a benign nevus. To explore whether alterations in the overall image color balance could influence the accuracy of skin cancer lesion classification by a CNN, we employed a differential evolution algorithm to search for subtle perturbations of global color balance that could lead to misclassification (Figure 2a).
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
