Calibration and Uncertainty for multiRater Volume Assessment in multiorgan Segmentation (CURVAS) challenge results

Deep learning (DL) has become the dominant approach for medical image segmentation, yet ensuring the reliability and clinical applicability of these models requires addressing key challenges such as annotation variability, calibration, and uncertainty estimation. This is why we created the Calibration and Uncertainty for multiRater Volume Assessment in multiorgan Segmentation (CURVAS), which highlights the critical role of multiple annotators in establishing a more comprehensive ground truth, emphasizing that segmentation is inherently subjective and that leveraging inter-annotator variability is essential for robust model evaluation. Seven teams participated in the challenge, submitting a variety of DL models evaluated using metrics such as Dice Similarity Coefficient (DSC), Expected Calibration Error (ECE), and Continuous Ranked Probability Score (CRPS). By incorporating consensus and dissensus ground truth, we assess how DL models handle uncertainty and whether their confidence estimates align with true segmentation performance. Our findings reinforce the importance of well-calibrated models, as better calibration is strongly correlated with the quality of the results. Furthermore, we demonstrate that segmentation models trained on diverse datasets and enriched with pre-trained knowledge exhibit greater robustness, particularly in cases deviating from standard anatomical structures. Notably, the best-performing models achieved high DSC and well-calibrated uncertainty estimates. This work underscores the need for multi-annotator ground truth, thorough calibration assessments, and uncertainty-aware evaluations to develop trustworthy and clinically reliable DL-based medical image segmentation models.

@article{riera-marin2025_2505.08685,
  title={ Calibration and Uncertainty for multiRater Volume Assessment in multiorgan Segmentation (CURVAS) challenge results },
  author={ Meritxell Riera-Marin and Sikha O K and Julia Rodriguez-Comas and Matthias Stefan May and Zhaohong Pan and Xiang Zhou and Xiaokun Liang and Franciskus Xaverius Erick and Andrea Prenner and Cedric Hemon and Valentin Boussot and Jean-Louis Dillenseger and Jean-Claude Nunes and Abdul Qayyum and Moona Mazher and Steven A Niederer and Kaisar Kushibar and Carlos Martin-Isla and Petia Radeva and Karim Lekadir and Theodore Barfoot and Luis C. Garcia Peraza Herrera and Ben Glocker and Tom Vercauteren and Lucas Gago and Justin Englemann and Joy-Marie Kleiss and Anton Aubanell and Andreu Antolin and Javier Garcia-Lopez and Miguel A. Gonzalez Ballester and Adrian Galdran },
  journal={arXiv preprint arXiv:2505.08685},
  year={ 2025 }
}