Practical uncertainty quantification for brain tumor segmentation

Moritz Fuchs, Camila González, Anirban Mukhopadhyay
Proceedings of The 5th International Conference on Medical Imaging with Deep Learning, PMLR 172:407-422, 2022.

Abstract

Despite U-Nets being the de-facto standard for medical image segmentation, researchers have identified shortcomings of U-Nets, such as overconfidence and poor out-of-distribution generalization. Several methods for uncertainty quantification try to solve such problems by relying on well-known approximations such as Monte-Carlo Drop-Out, Probabilistic U-Net, and Stochastic Segmentation Networks. We introduce a novel multi-headed Variational U-Net. The proposed approach combines the global exploration capabilities of deep ensembles with the out-of-distribution robustness of Variational Inference. An efficient training strategy and an expressive yet general design ensure superior uncertainty quantification within a reasonable compute requirement. We thoroughly analyze the performance and properties of our approach on the publicly available BRATS2018 dataset. Further, we test our model on four commonly observed distribution shifts. The proposed approach has good uncertainty calibration and is robust to out-of-distribution shifts.

Cite this Paper

BibTeX
@InProceedings{pmlr-v172-fuchs22a, title = {Practical uncertainty quantification for brain tumor segmentation}, author = {Fuchs, Moritz and Gonz\'{a}lez, Camila and Mukhopadhyay, Anirban}, booktitle = {Proceedings of The 5th International Conference on Medical Imaging with Deep Learning}, pages = {407--422}, year = {2022}, editor = {Konukoglu, Ender and Menze, Bjoern and Venkataraman, Archana and Baumgartner, Christian and Dou, Qi and Albarqouni, Shadi}, volume = {172}, series = {Proceedings of Machine Learning Research}, month = {06--08 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v172/fuchs22a/fuchs22a.pdf}, url = {https://proceedings.mlr.press/v172/fuchs22a.html}, abstract = {Despite U-Nets being the de-facto standard for medical image segmentation, researchers have identified shortcomings of U-Nets, such as overconfidence and poor out-of-distribution generalization. Several methods for uncertainty quantification try to solve such problems by relying on well-known approximations such as Monte-Carlo Drop-Out, Probabilistic U-Net, and Stochastic Segmentation Networks. We introduce a novel multi-headed Variational U-Net. The proposed approach combines the global exploration capabilities of deep ensembles with the out-of-distribution robustness of Variational Inference. An efficient training strategy and an expressive yet general design ensure superior uncertainty quantification within a reasonable compute requirement. We thoroughly analyze the performance and properties of our approach on the publicly available BRATS2018 dataset. Further, we test our model on four commonly observed distribution shifts. The proposed approach has good uncertainty calibration and is robust to out-of-distribution shifts.} }
Endnote
%0 Conference Paper %T Practical uncertainty quantification for brain tumor segmentation %A Moritz Fuchs %A Camila González %A Anirban Mukhopadhyay %B Proceedings of The 5th International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2022 %E Ender Konukoglu %E Bjoern Menze %E Archana Venkataraman %E Christian Baumgartner %E Qi Dou %E Shadi Albarqouni %F pmlr-v172-fuchs22a %I PMLR %P 407--422 %U https://proceedings.mlr.press/v172/fuchs22a.html %V 172 %X Despite U-Nets being the de-facto standard for medical image segmentation, researchers have identified shortcomings of U-Nets, such as overconfidence and poor out-of-distribution generalization. Several methods for uncertainty quantification try to solve such problems by relying on well-known approximations such as Monte-Carlo Drop-Out, Probabilistic U-Net, and Stochastic Segmentation Networks. We introduce a novel multi-headed Variational U-Net. The proposed approach combines the global exploration capabilities of deep ensembles with the out-of-distribution robustness of Variational Inference. An efficient training strategy and an expressive yet general design ensure superior uncertainty quantification within a reasonable compute requirement. We thoroughly analyze the performance and properties of our approach on the publicly available BRATS2018 dataset. Further, we test our model on four commonly observed distribution shifts. The proposed approach has good uncertainty calibration and is robust to out-of-distribution shifts.
APA
Fuchs, M., González, C. & Mukhopadhyay, A.. (2022). Practical uncertainty quantification for brain tumor segmentation. Proceedings of The 5th International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 172:407-422 Available from https://proceedings.mlr.press/v172/fuchs22a.html.

Related Material