Vasudevan Lakshminarayanan

Unmasking clinical shortcuts: a latent-space audit of dense convolutional networks for chest radiography

Deep convolutional neural networks (CNNs) in medical imaging can achieve acceptable global discrimination while still relying on shortcut-sensitive signals that undermine reliability in clinically important subgroups. In chest radiography, non-pathological factors such as image acquisition view, support devices, and su...