Understanding and Visualizing Generalization in UNets
Published in Medical Imaging with Deep Learning (MIDL) 2021 in Lübeck, Germany, 2021
Recommended citation: Rajagopal, A., Madala, V.C., Hope, T.A.; Larson, P.. (2021). Understanding and Visualizing Generalization in UNets. *Proceedings of the Fourth Conference on Medical Imaging with Deep Learning*, in *Proceedings of Machine Learning Research* 143:665-681 Available from https://proceedings.mlr.press/v143/rajagopal21a.html. https://2021.midl.io/proceedings/rajagopal21.pdf
Fully-convolutional neural networks, such as 2D and 3D UNets, are now pervasive in medi- cal imaging for semantic segmentation, classification, image denoising, domain translation, and reconstruction. However, evaluation of UNet performance, as with most CNNs, has mostly been relegated to evaluation of a few performance metrics (e.g. accuracy, IoU, SSIM, etc.) using the network’s final predictions, which provides little insight into impor- tant issues such as generalization and dataset shift that can occur in clinical applications. In this paper, we propose techniques for understanding and visualizing the generalization performance of UNets in image classification and regression tasks, as well as metrics that are indicative of performance on unseen, unlabeled data.
Download paper here: Link1
Recommended citation: ‘Rajagopal, A., Madala, V.C., Hope, T.A.; Larson, P.. (2021). Understanding and Visualizing Generalization in UNets. Proceedings of the Fourth Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 143:665-681 Available from https://proceedings.mlr.press/v143/rajagopal21a.html.’