mlsanityData to 7 Oct 2026

A Quantitative Comparison of Epistemic Uncertainty Maps Applied to Multi-Class Segmentation

arXiv 2109.10702 ↗First posted 22 Sep 2021cs.CV eess.IVcited by 9

Not on the map. Why?

Abstract (arXiv)

Uncertainty assessment has gained rapid interest in medical image analysis. A popular technique to compute epistemic uncertainty is the Monte-Carlo (MC) dropout technique. From a network with MC dropout and a single input, multiple outputs can be sampled. Various methods can be used to obtain epistemic uncertainty maps from those multiple outputs. In the case of multi-class segmentation, the number of methods is even larger as epistemic uncertainty can be computed voxelwise per class or voxelwise per image. This paper highlights a systematic approach to define and quantitatively compare those methods in two different contexts: class-specific epistemic uncertainty maps (one value per image, voxel and class) and combined epistemic uncertainty maps (one value per image and voxel). We applied this quantitative analysis to a multi-class segmentation of the carotid artery lumen and vessel wall, on a multi-center, multi-scanner, multi-sequence dataset of (MR) images. We validated our analysis over 144 sets of hyperparameters of a model. Our main analysis considers the relationship between the order of the voxels sorted according to their epistemic uncertainty values and the misclassification of the prediction. Under this consideration, the comparison of combined uncertainty maps reveals that the multi-class entropy and the multi-class mutual information statistically out-perform the other combined uncertainty maps under study. In a class-specific scenario, the one-versus-all entropy statistically out-performs the class-wise entropy, the class-wise variance and the one versus all mutual information. The class-wise entropy statistically out-performs the other class-specific uncertainty maps in terms of calibration. We made a python package available to reproduce our analysis on different data and tasks.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Builds on 25 of its 53 references that are in the map. First posted on arXiv 22 Sep 2021.

Builds on

25 of 53 references.

References · 25

  1. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  2. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  3. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,416
  4. 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,888
  5. 2017What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?cited by 2,335
  6. 2016V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentationcited by 2,284
  7. 2018Averaging Weights Leads to Wider Optima and Better Generalizationcited by 1,194
  8. 2020A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challengescited by 813
  9. 2019A Simple Baseline for Bayesian Uncertainty in Deep Learningcited by 555
  10. 2020Bayesian Deep Learning and a Probabilistic Perspective of Generalizationcited by 480
  11. 2018Aleatoric uncertainty estimation with test-time augmentation for medical image segmentation with convolutional neural networkscited by 278
  12. 2015Bayesian SegNet: Model Uncertainty in Deep Convolutional Encoder-Decoder Architectures for Scene Understandingcited by 218
13 more
  1. 2019Evaluating Scalable Bayesian Deep Learning Methods for Robust Computer Visioncited by 182
  2. 2019Confidence Calibration and Predictive Uncertainty Estimation for Deep Medical Image Segmentationcited by 162
  3. 2018Exploring Uncertainty Measures in Deep Networks for Multiple Sclerosis Lesion Detection and Segmentationcited by 146
  4. 2018Evaluating Bayesian Deep Learning Methods for Semantic Segmentationcited by 143
  5. 2019Deep Semantic Segmentation of Natural and Medical Images: A Reviewcited by 124
  6. 2019The Fishyscapes Benchmark: Measuring Blind Spots in Semantic Segmentationcited by 122
  7. 2019DropConnect Is Effective in Modeling Uncertainty of Bayesian Deep Networkscited by 92
  8. 2018Bayesian Uncertainty Estimation for Batch Normalized Deep Networkscited by 70
  9. 2018Evaluating Uncertainty Quantification in End-to-End Autonomous Driving Controlcited by 54
  10. 2018On the Effect of Inter-observer Variability for a Reliable Estimation of Uncertainty of Medical Image Segmentationcited by 46
  11. 2019Exploiting Epistemic Uncertainty of Anatomy Segmentation for Anomaly Detection in Retinal OCTcited by 41
  12. 2020Uncertainty Evaluation Metric for Brain Tumour Segmentationcited by 11
  13. 2019U2-Net: A Bayesian U-Net model with epistemic uncertainty feedback for photoreceptor layer segmentation in pathological OCT scanscited by 9

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 3; 2024: 2; 2025: 1; 2026: 2.

9 papers.

  1. 2023A review of uncertainty quantification in medical image analysis: probabilistic and non-probabilistic methodscited by 39
  2. 2023Anatomically-aware Uncertainty for Semi-supervised Image Segmentationcited by 13
  3. 2024A Review of Bayesian Uncertainty Quantification in Deep Probabilistic Image Segmentationcited by 6
  4. 2023How inter-rater variability relates to aleatoric and epistemic uncertainty: a case study with deep learning-based paraspinal muscle segmentationcited by 4
  5. 2022Influence of uncertainty estimation techniques on false-positive reduction in liver lesion detectioncited by 3
  6. 2025ConfIC-RCA: Statistically Grounded Efficient Estimation of Segmentation Qualitycited by 3
  7. 2026Not Just How Much, But Where: Decomposing Epistemic Uncertainty into Per-Class Contributionscited by 2
  8. 2024Improving Uncertainty-Error Correspondence in Deep Bayesian Medical Image Segmentationcited by 1
  9. 2026Better than Average: Spatially-Aware Aggregation of Segmentation Uncertainty Improves Downstream Performancecited by 1