Adversarial Attacks on Dental Diseases’ Classifiers

Abstract

With the development of deep learning models, they have found strong foothold in medicinal practices. Dentists are relying more and more on deep learning classification models. However, the credibility of these models has been questioned owing to the discovery of adversarial attacks on them. These attacks manipulate images leading them to be misclassified by state-of-the-art DNN models. This is why it is pertinent to study about the impact of adversarial attacks on deep learning classifiers, particularly in medical domain.