Informative sample generation using class aware generative adversarial networks for classification of chest Xrays

Behzad Bozorgtabar, Dwarikanath Mahapatra, Hendrik von Teng, Alexander Pollinger, Lukas Ebner, Jean-Philippe Thiran, Mauricio Reyes

Abstract: Highlights•A GAN based active learning method is proposed for Xray image classification.•By informative sample selection we obtain state of the art results with fewer samples.•Fewer labeled samples translates to less effort,shallower networks and less resources.