Improving confidence in MRI-based auto-segmentation via uncertainty assessment
DOI:
https://doi.org/10.2340/1651-226X.2026.45685Keywords:
deep learning segmentation, brain magnetic resonance imaging, uncertainty, Brain neoplasms, deep learningAbstract
Background and purpose: Accurate delineation of organs of interest (OOIs, also commonly referred to as organs at risk, OARs) is crucial for safe radiotherapy. While deep learning-based segmentation using convolutional neural networks has achieved high geometric accuracy, clinical translation is hindered by overconfident, uncalibrated predictions in anatomically ambiguous regions. Uncertainty quantification and model calibration are prerequisites for safe clinical workflows. This study compared the standard nnU-Netv2 against its residual-encoding variant (ResEncM), hypothesizing that ResEncM would demonstrate superior reliability and calibration while maintaining comparable geometric accuracy.
Patient/material and methods: T1-weighted contrast-enhanced MRI scans from 70 brain cancer patients were used (55 training/validation, 15 testing). Ground-truth contours for brainstem, hippocampi, chiasm, optic nerves, optic tracts, and pituitary were delineated per Danish Neuro Oncology Group guidelines. Both architectures were trained using five-fold cross-validation with identical preprocessing. Epistemic uncertainty was quantified using mutual information, and Expected Calibration Error (ECE) was computed within a 10-mm isotropic margin around reference contours.
Results: Both models achieved high geometric accuracy (brainstem dice similarity coefficient [DSC] > 0.93, hippocampi DSC > 0.81). No significant geometric differences were found for large structures. ResEncM showed significantly lower DSC for the pituitary (p = 0.003) and chiasm (p = 0.018). However, ResEncM demonstrated significantly lower epistemic uncertainty and ensemble variance across all structures
(p < 0.05), and significantly reduced ECE for the optic chiasm, optic tracts, and pituitary.
Interpretation: Integrating a deep residual encoder into the standard U-Net framework significantly improves reliability and calibration of automated brain OOI contours while maintaining strong geometric performance. The ResEncM architecture provides a more trustworthy tool for clinical radiotherapy by reliably flagging high-uncertainty voxels, supporting confidence-aware clinical workflows.
Downloads
References
Lorenzen EL, Kallehauge JF, Byskov CS, Dahlrot RH, Haslund CA, Guldberg TL, et al. A national study on the inter-observer variability in the delineation of organs at risk in the brain. Acta Oncol. 2021;60(11):1548–54. DOI: https://doi.org/10.1080/0284186X.2021.1975813
Ronneberger O, Fischer P, Brox T. U-Net: convolutional networks for biomedical image segmentation. CoRR. 2015;abs/1505.04597:9351. DOI: https://doi.org/10.1007/978-3-319-24574-4_28
Isensee F, Jaeger PF, Kohl SA, Petersen J, Maier-Hein KH. nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation. Nat Methods. 2021;18(2):203–11. DOI: https://doi.org/10.1038/s41592-020-01008-z
Wahid KA, Ahmed S, Glerean E, Heukelom J, van der Heide UA, Fuller CD, et al. Artificial intelligence uncertainty quantification in radiotherapy applications – a scoping review. Radiother Oncol. 2024;201:110542. DOI: https://doi.org/10.1016/j.radonc.2024.110542
Korreman SS, Ren J. Understanding and leveraging uncertainties in autosegmentation for radiotherapy. BJR AI. 2025;2(1):ubaf013. DOI: https://doi.org/10.1093/bjrai/ubaf013
Hüllermeier E, Waegeman W. Aleatoric and epistemic uncertainty in machine learning: an introduction to concepts and methods. Mach Learn. 2021;110:457–506. DOI: https://doi.org/10.1007/s10994-021-05946-3
Ovadia Y, Fertig E, Ren J, Nado Z, Sculley D, Nowozin S, et al. Can you trust your model’s uncertainty? Evaluating predictive uncertainty under dataset shift. arXiv. 2019. [cited 2026 May 1]. Available from: https://arxiv.org/abs/1906.02530
Mehrtash A, Wells WM, Tempany CM, Abolmaesumi P, Kapur T. Confidence calibration and predictive uncertainty estimation for deep medical image segmentation. IEEE Trans Med Imaging. 2020;39(12):3868–78. DOI: https://doi.org/10.1109/TMI.2020.3006437
Isensee F, Wald T, Ulrich C, Baumgartner M, Roy S, Maier-Hein K, et al. nnU-Net revisited: a call for rigorous validation in 3D medical image segmentation. arXiv. 2024. [cited 2026 May 1]. Available from: https://arxiv.org/abs/2404.09556 DOI: https://doi.org/10.1007/978-3-031-72114-4_47
Danish Neuro Oncology Group (DNOG) guidelines. [cited 2026 May 1]. Available from: https://dnog.dk/onewebmedia/files/Retningslinier%20PDF/DNOG%202023%20Retningslinjer%20for%20straalebehandling.pdf
Turcas A, Leucuta D, Balan C, Clementel E, Gheara C, Kacso A, et al. Deep-learning magnetic resonance imaging-based automatic segmentation for organs-at-risk in the brain: accuracy and impact on dose distribution. Phys Imaging Radiat Oncol. 2023;27:100454. DOI: https://doi.org/10.1016/j.phro.2023.100454
Ren J, Teuwen J, Nijkamp J, Rasmussen M, Gouw Z, Eriksen JG, et al. Enhancing the reliability of deep learning-based head and neck tumour segmentation using uncertainty estimation with multi-modal images. Phys Med Biol. 2024;69(16):165018. DOI: https://doi.org/10.1088/1361-6560/ad682d
Ghaffari M, Sowmya A, Oliver R. Automated brain tumour segmentation using cascaded 3D densely-connected U-Net. In: Crimi A, Bakas S, editors. Brainlesion: glioma, multiple sclerosis, stroke and traumatic brain injuries. BrainLes 2020. Cham: Springer; 2021. p. 467–77. DOI: https://doi.org/10.1007/978-3-030-72084-1_43
Kumar A, Ghosal P, Kundu SS, Mukherjee A, Nandi D. A lightweight asymmetric U-Net framework for acute ischemic stroke lesion segmentation in CT and CTP images. Comput Methods Programs Biomed. 2022;226:107157. DOI: https://doi.org/10.1016/j.cmpb.2022.107157
Additional Files
Published
How to Cite
License
Copyright (c) 2026 Jesper Folsted Kallehauge, Jintao Ren, Yasmin Lassen-Ramshad
This work is licensed under a Creative Commons Attribution 4.0 International License.
