TY - GEN
T1 - Uncertainty quantification in medical image segmentation with normalizing flows
AU - Selvan, Raghavendra
AU - Faye, Frederik
AU - Middleton, Jon
AU - Pai, Akshay
N1 - Accepted to be presented at 11th International Workshop on Machine Learning in Medical Imaging. Source code will be updated at https://github.com/raghavian/cFlow
PY - 2020
Y1 - 2020
N2 - Medical image segmentation is inherently an ambiguous task due to factors such as partial volumes and variations in anatomical definitions. While in most cases the segmentation uncertainty is around the border of structures of interest, there can also be considerable inter-rater differences. The class of conditional variational autoencoders (cVAE) offers a principled approach to inferring distributions over plausible segmentations that are conditioned on input images. Segmentation uncertainty estimated from samples of such distributions can be more informative than using pixel level probability scores. In this work, we propose a novel conditional generative model that is based on conditional Normalizing Flow (cFlow). The basic idea is to increase the expressivity of the cVAE by introducing a cFlow transformation step after the encoder. This yields improved approximations of the latent posterior distribution, allowing the model to capture richer segmentation variations. With this we show that the quality and diversity of samples obtained from our conditional generative model is enhanced. Performance of our model, which we call cFlow Net, is evaluated on two medical imaging datasets demonstrating substantial improvements in both qualitative and quantitative measures when compared to a recent cVAE based model.
AB - Medical image segmentation is inherently an ambiguous task due to factors such as partial volumes and variations in anatomical definitions. While in most cases the segmentation uncertainty is around the border of structures of interest, there can also be considerable inter-rater differences. The class of conditional variational autoencoders (cVAE) offers a principled approach to inferring distributions over plausible segmentations that are conditioned on input images. Segmentation uncertainty estimated from samples of such distributions can be more informative than using pixel level probability scores. In this work, we propose a novel conditional generative model that is based on conditional Normalizing Flow (cFlow). The basic idea is to increase the expressivity of the cVAE by introducing a cFlow transformation step after the encoder. This yields improved approximations of the latent posterior distribution, allowing the model to capture richer segmentation variations. With this we show that the quality and diversity of samples obtained from our conditional generative model is enhanced. Performance of our model, which we call cFlow Net, is evaluated on two medical imaging datasets demonstrating substantial improvements in both qualitative and quantitative measures when compared to a recent cVAE based model.
KW - stat.ML
KW - cs.CV
KW - cs.LG
U2 - 10.1007/978-3-030-59861-7_9
DO - 10.1007/978-3-030-59861-7_9
M3 - Article in proceedings
T3 - Lecture Notes in Computer Science
BT - Machine Learning in Medical Imaging
PB - Springer
T2 - 11th International Workshop on Machine Learning in Medical Imaging, MLMI 2020, held in conjunction with the 23rd International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2020
Y2 - 4 October 2020 through 4 October 2020
ER -