StyleGAN2-based Out-of-Distribution Detection for Medical Imaging
This addresses the problem of unreliable deep learning models in clinical settings due to out-of-distribution images, but it is incremental as it applies an existing GAN method to medical data.
The paper tackled detecting out-of-distribution medical images using a StyleGAN2-based method, achieving over 90% AUROC for distinguishing liver from non-liver CT scans but failing to reconstruct liver artifacts like needles and ascites.
One barrier to the clinical deployment of deep learning-based models is the presence of images at runtime that lie far outside the training distribution of a given model. We aim to detect these out-of-distribution (OOD) images with a generative adversarial network (GAN). Our training dataset was comprised of 3,234 liver-containing computed tomography (CT) scans from 456 patients. Our OOD test data consisted of CT images of the brain, head and neck, lung, cervix, and abnormal livers. A StyleGAN2-ADA architecture was employed to model the training distribution. Images were reconstructed using backpropagation. Reconstructions were evaluated using the Wasserstein distance, mean squared error, and the structural similarity index measure. OOD detection was evaluated with the area under the receiver operating characteristic curve (AUROC). Our paradigm distinguished between liver and non-liver CT with greater than 90% AUROC. It was also completely unable to reconstruct liver artifacts, such as needles and ascites.