Spatially Varying Label Smoothing: Capturing Uncertainty from Expert Annotations
arXiv:2104.05788
Abstract
The task of image segmentation is inherently noisy due to ambiguities regarding the exact location of boundaries between anatomical structures. We argue that this information can be extracted from the expert annotations at no extra cost, and when integrated into state-of-the-art neural networks, it can lead to improved calibration between soft probabilistic predictions and the underlying uncertainty. We built upon label smoothing (LS) where a network is trained on 'blurred' versions of the ground truth labels which has been shown to be effective for calibrating output predictions. However, LS is not taking the local structure into account and results in overly smoothed predictions with low confidence even for non-ambiguous regions. Here, we propose Spatially Varying Label Smoothing (SVLS), a soft labeling technique that captures the structural uncertainty in semantic segmentation. SVLS also naturally lends itself to incorporate inter-rater uncertainty when multiple labelmaps are available. The proposed approach is extensively validated on four clinical segmentation tasks with different imaging modalities, number of classes and single and multi-rater expert annotations. The results demonstrate that SVLS, despite its simplicity, obtains superior boundary prediction with improved uncertainty and model calibration.
Accepted at IPMI 2021
References in corpus (10)
- On Calibration of Modern Neural Networks
- When Does Label Smoothing Help?
- The KiTS19 Challenge Data: 300 Kidney Tumor Cases with Clinical Context, CT Semantic Segmentations, and Surgical Outcomes
- Deep learning to achieve clinically applicable segmentation of head and neck anatomy for radiotherapy
- Uncertainty and Interpretability in Convolutional Neural Networks for Semantic Segmentation of Colorectal Polyps
- Beyond temperature scaling: Obtaining well-calibrated multiclass probabilities with Dirichlet calibration
- Stochastic Segmentation Networks: Modelling Spatially Correlated Aleatoric Uncertainty
- Towards better decoding and language model integration in sequence to sequence models
- Dropout as a Bayesian Approximation: Appendix
- Well-calibrated Model Uncertainty with Temperature Scaling for Dropout Variational Inference