Title
Efficient Bayesian Uncertainty Estimation for nnU-Net
Author
Zhao, Y. (TU Delft ImPhys/Medical Imaging)
Yang, C. (TU Delft ImPhys/Medical Imaging)
Schweidtmann, A.M. (TU Delft ChemE/Product and Process Engineering)
Tao, Q. (TU Delft ImPhys/Medical Imaging)
Contributor
Wang, Linwei (editor)
Dou, Qi (editor)
Fletcher, P. Thomas (editor)
Speidel, Stefanie (editor)
Li, Shuo (editor)
Date
2022
Abstract
The self-configuring nnU-Net has achieved leading performance in a large range of medical image segmentation challenges. It is widely considered as the model of choice and a strong baseline for medical image segmentation. However, despite its extraordinary performance, nnU-Net does not supply a measure of uncertainty to indicate its possible failure. This can be problematic for large-scale image segmentation applications, where data are heterogeneous and nnU-Net may fail without notice. In this work, we introduce a novel method to estimate nnU-Net uncertainty for medical image segmentation. We propose a highly effective scheme for posterior sampling of weight space for Bayesian uncertainty estimation. Different from previous baseline methods such as Monte Carlo Dropout and mean-field Bayesian Neural Networks, our proposed method does not require a variational architecture and keeps the original nnU-Net architecture intact, thereby preserving its excellent performance and ease of use. Additionally, we boost the segmentation performance over the original nnU-Net via marginalizing multi-modal posterior models. We applied our method on the public ACDC and M &M datasets of cardiac MRI and demonstrated improved uncertainty estimation over a range of baseline methods. The proposed method further strengthens nnU-Net for medical image segmentation in terms of both segmentation accuracy and quality control.
Subject
nnU-Net
Stochastic gradient descent
Uncertainty estimation
Variational inference
To reference this document use:
http://resolver.tudelft.nl/uuid:938b8ec5-13ab-4f90-801b-529cc8608d66
DOI
https://doi.org/10.1007/978-3-031-16452-1_51
Publisher
Springer
Embargo date
2023-07-01
ISBN
978-3-031-16451-4
Source
Medical Image Computing and Computer Assisted Intervention – MICCAI 2022 - 25th International Conference, Proceedings
Event
25th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2022, 2022-09-18 → 2022-09-22, Singapore, Singapore
Series
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 0302-9743, 13438 LNCS
Bibliographical note
Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.
Part of collection
Institutional Repository
Document type
conference paper
Rights
© 2022 Y. Zhao, C. Yang, A.M. Schweidtmann, Q. Tao