Biblio

In recent years, the security of AI systems has drawn increasing research attention, especially in the medical imaging realm. To develop a secure medical image analysis (MIA) system, it is a must to study possible backdoor attacks (BAs), which can embed hidden malicious behaviors into the system. However, designing a unified BA method that can be applied to various MIA systems is challenging due to the diversity of imaging modalities (e.g., X-Ray, CT, and MRI) and analysis tasks (e.g., classification, detection, and segmentation). Most existing BA methods are designed to attack natural image classification models, which apply spatial triggers to training images and inevitably corrupt the semantics of poisoned pixels, leading to the failures of attacking dense prediction models. To address this issue, we propose a novel Frequency-Injection based Backdoor Attack method (FIBA) that is capable of delivering attacks in various MIA tasks. Specifically, FIBA leverages a trigger function in the frequency domain that can inject the low-frequency information of a trigger image into the poisoned image by linearly combining the spectral amplitude of both images. Since it preserves the semantics of the poisoned image pixels, FIBA can perform attacks on both classification and dense prediction models. Experiments on three benchmarks in MIA (i.e., ISIC-2019 [4] for skin lesion classification, KiTS-19 [17] for kidney tumor segmentation, and EAD-2019 [1] for endoscopic artifact detection), validate the effectiveness of FIBA and its superiority over stateof-the-art methods in attacking MIA models and bypassing backdoor defense. Source code will be available at code.

We introduce AdaMix, an adaptive differentially private algorithm for training deep neural network classifiers using both private and public image data. While pre-training language models on large public datasets has enabled strong differential privacy (DP) guarantees with minor loss of accuracy, a similar practice yields punishing trade-offs in vision tasks. A few-shot or even zero-shot learning baseline that ignores private data can outperform fine-tuning on a large private dataset. AdaMix incorporates few-shot training, or cross-modal zero-shot learning, on public data prior to private fine-tuning, to improve the trade-off. AdaMix reduces the error increase from the non-private upper bound from the 167–311% of the baseline, on average across 6 datasets, to 68-92% depending on the desired privacy level selected by the user. AdaMix tackles the trade-off arising in visual classification, whereby the most privacy sensitive data, corresponding to isolated points in representation space, are also critical for high classification accuracy. In addition, AdaMix comes with strong theoretical privacy guarantees and convergence analysis.