mlsanityData to 7 Oct 2026

Average Calibration Error: A Differentiable Loss for Improved Reliability in Image Segmentation

arXiv 2403.06759 ↗First posted 11 Mar 2024cs.CV cs.LGcited by 6

Not on the map. Why?

Abstract (arXiv)

Deep neural networks for medical image segmentation often produce overconfident results misaligned with empirical observations. Such miscalibration, challenges their clinical translation. We propose to use marginal L1 average calibration error (mL1-ACE) as a novel auxiliary loss function to improve pixel-wise calibration without compromising segmentation quality. We show that this loss, despite using hard binning, is directly differentiable, bypassing the need for approximate but differentiable surrogate or soft binning approaches. Our work also introduces the concept of dataset reliability histograms which generalises standard reliability diagrams for refined visual assessment of calibration in semantic segmentation aggregated at the dataset level. Using mL1-ACE, we reduce average and maximum calibration error by 45% and 55% respectively, maintaining a Dice score of 87% on the BraTS 2021 dataset. We share our code here: https://github.com/cai4cai/ACE-DLIRIS

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 15 of its 24 references that are in the map. First posted on arXiv 11 Mar 2024.

Builds on

15 of 24 references.

References · 15

  1. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  2. 2017On Calibration of Modern Neural Networkscited by 4,860
  3. 2021The RSNA-ASNR-MICCAI BraTS 2021 Benchmark on Brain Tumor Segmentation and Radiogenomic Classificationcited by 507
  4. 2022MONAI: An open-source framework for deep learning in healthcarecited by 438
  5. 2019Beyond temperature scaling: Obtaining well-calibrated multiclass probabilities with Dirichlet calibrationcited by 359
  6. 2019Confidence Calibration and Predictive Uncertainty Estimation for Deep Medical Image Segmentationcited by 162
  7. 2020Local Temperature Scaling for Probability Calibrationcited by 95
  8. 2022A Stitch in Time Saves Nine: A Train-Time Regularizing Loss for Improved Neural Network Calibrationcited by 49
  9. 2020Improved Trainable Calibration Method for Neural Networks on Medical Imaging Classificationcited by 37
  10. 2021Meta-Calibration: Learning of Model Calibration Using Differentiable Expected Calibration Errorcited by 28
  11. 2021Class-Distribution-Aware Calibration for Long-Tailed Visual Recognitioncited by 19
  12. 2021Calibrating the Dice loss to handle neural network overconfidence for biomedical image segmentationcited by 17
3 more
  1. 2023Trust your neighbours: Penalty-based constraints for model calibrationcited by 12
  2. 2024Neighbor-Aware Calibration of Segmentation Networks with Penalty-Based Constraintscited by 7
  3. 2021Generalized Wasserstein Dice Loss, Test-time Augmentation, and Transformers for the BraTS 2021 challengecited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 1; 2025: 4; 2026: 1.

6 papers.

  1. 2025Calibration and Uncertainty for multiRater Volume Assessment in multiorgan Segmentation (CURVAS) challenge resultscited by 7
  2. 2025Average Calibration Losses for Reliable Uncertainty in Medical Image Segmentationcited by 3
  3. 2025Flow Stochastic Segmentation Networkscited by 3
  4. 2024OOD-SEG: Exploiting out-of-distribution detection techniques for learning image segmentation from sparse multi-class positive-only annotationscited by 2
  5. 2025Rethinking Glaucoma Calibration: Voting-Based Binocular and Metadata Integrationcited by 0
  6. 2026Beyond scalar losses: calibrating segmentation models via gradient vector field surgerycited by 0