IEEE

Minimizing Adversarial Training Samples for Robust Image Classifiers: Analysis and Adversarial Example Generator Design

Abstract: Training deep neural networks (DNNs) with altered data, known as adversarial training, is essential for improving their robustness. A significant challenge emerges as the robustness ...
IEEE

NSB-H 2 GAN: “Negative Sample”-Boosted Hierarchical Heterogeneous Graph Attention Network for Interpretable Classification of Whole-Slide Images

Abstract: Gigapixel whole-slide image (WSI) prediction and region-of-interest localization present considerable challenges due to the diverse range of features both across different slides and within ...
ahajournals.org

Abstract A053: Development and multicentric validation of a publicly-available head CT deidentification tool to facilitate inter-institutional stroke image sharing

Purpose: Head CTs of stroke patients contain personal health information (PHI) in Digital Imaging and Communications in Medicine (DICOM) metadata and facial features in 3D reconstruction, raising ...