Adversarially Robust Feature Learning for Breast Cancer Diagnosis
This work addresses adversarial robustness in breast cancer diagnosis, a critical safety issue in clinical settings, though it is incremental as it builds on existing adversarial training techniques.
The authors tackled the problem of adversarial attacks in deep learning for breast cancer diagnosis by proposing an adversarially robust feature learning method, which outperformed state-of-the-art methods on two clinical datasets totaling 9,548 mammogram images.
Adversarial data can lead to malfunction of deep learning applications. It is essential to develop deep learning models that are robust to adversarial data while accurate on standard, clean data. In this study, we proposed a novel adversarially robust feature learning (ARFL) method for a real-world application of breast cancer diagnosis. ARFL facilitates adversarial training using both standard data and adversarial data, where a feature correlation measure is incorporated as an objective function to encourage learning of robust features and restrain spurious features. To show the effects of ARFL in breast cancer diagnosis, we built and evaluated diagnosis models using two independent clinically collected breast imaging datasets, comprising a total of 9,548 mammogram images. We performed extensive experiments showing that our method outperformed several state-of-the-art methods and that our method can enhance safer breast cancer diagnosis against adversarial attacks in clinical settings.