Reliable brain tumor segmentation using calibSeg: A temperature-calibrated 3D Attention-UNet with uncertainty quantification
Accurate brain-tumor segmentation from multi-modal magnetic resonance imaging is essential for diagnosis, treatment planning, and longitudinal assessment, but high segmentation accuracy alone does not ensure reliable model confidence. This study proposes CalibSeg, a temperature-calibrated three-dimensional Attention-UNet that integrates attention-guided feature selection, Monte Carlo dropout-based uncertainty estimation, and post-hoc temperature scaling. The framework was evaluated on the Brain Tumor Segmentation 2020 benchmark using T1, contrast-enhanced T1, T2, and FLAIR magnetic resonance imaging modalities. Based on the reported performance metrics, CalibSeg achieved 95.84% precision, 100.00% recall, 100.00% specificity, 95.89% accuracy, 97.86% F1-score, and 97.26% overall accuracy. The calibration analysis showed relative reductions of 23% in expected calibration error and 19% in maximum calibration error after temperature scaling. Monte Carlo dropout entropy provides a complementary uncertainty signal. CalibSeg performed better than the three baseline implementations evaluated in this study; however, the evidence is limited to aggregate metrics from a single public dataset, without confidence intervals, repeated-run variability, per-case segmentation distributions, direct comparison with recent transformer-based models, or external clinical validation. Accordingly, the findings should be interpreted as preliminary evidence of improved baseline performance and reliability-aware prediction rather than proof of comprehensive state-of-the-art superiority or clinical reliability.
Abraham, N., & Khan, N. M. (2019). A novel focal Tversky loss function with improved Attention U-Net for lesion segmentation. In 2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019) (pp. 683–687). IEEE. https://doi.org/10.1109/ISBI.2019.8759329
Brunekreef, J., Marcus, E., Sheombarsing, R., Sonke, J. J., & Teuwen, J. (2024). Kandinsky conformal prediction: Efficient calibration of image segmentation algorithms. In 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (pp. 4135–4143). IEEE. https://doi.org/10.1109/CVPR52733.2024.00396
Cao, H., Wang, Y., Chen, J., Jiang, D., Zhang, X., Tian, Q., & Wang, M. (2021). Swin-Unet: Unet-like pure transformer for medical image segmentation (Version 1). arXiv. https://doi.org/10.48550/ARXIV.2105.05537
Chen, J., Lu, Y., Yu, Q., Luo, X., Adeli, E., Wang, Y., Lu, L., Yuille, A. L., & Zhou, Y. (2021). TransUNet: Transformers make strong encoders for medical image segmentation (Version 1). arXiv. https://doi.org/10.48550/ARXIV.2102.04306
Çiçek, Ö., Abdulkadir, A., Lienkamp, S. S., Brox, T., & Ronneberger, O. (2016). 3D U-Net: Learning dense volumetric segmentation from sparse annotation. In Lecture Notes in Computer Science (pp. 424–432). Springer International Publishing, Cham, Switzerland. https://doi.org/10.1007/978-3-319-46723-8_49
Hatamizadeh, A., Nath, V., Tang, Y., Yang, D., Roth, H. R., & Xu, D. (2022a). Swin UNETR: Swin transformers for semantic segmentation of brain tumors in MRI images (Version 1). arXiv. https://doi.org/10.48550/arXiv.2201.01266
Hatamizadeh, A., Tang, Y., Nath, V., Yang, D., Myronenko, A., Landman, B., Roth, H. R., & Xu, D. (2022b). UNETR: Transformers for 3D medical image segmentation. In 2022 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) (pp. 1748–1758). IEEE. https://doi.org/10.1109/WACV51458.2022.00181
Havaei, M., Davy, A., Warde-Farley, D., Biard, A., Courville, A., Bengio, Y., Pal, C., Jodoin, P.-M., & Larochelle, H. (2017). Brain tumor segmentation with deep neural networks. Medical Image Analysis, 35, 18–31. https://doi.org/10.1016/j.media.2016.05.004
Isensee, F., Jaeger, P. F., Kohl, S. A. A., Petersen, J., & Maier-Hein, K. H. (2021). nnU-Net: A self-configuring method for deep learning-based biomedical image segmentation. Nature Methods, 18(2), 203–211. https://doi.org/10.1038/s41592-020-01008-z
Kendall, A., & Gal, Y. (2017). What uncertainties do we need in Bayesian deep learning for computer vision? Advances in Neural Information Processing Systems, 30.
Lin, T. Y., Goyal, P., Girshick, R., He, K., & Dollar, P. (2017). Focal loss for dense object detection. In 2017 IEEE International Conference on Computer Vision (ICCV) (pp. 2999–3007). IEEE. https://doi.org/10.1109/ICCV.2017.324
Menze, B. H., Jakab, A., Bauer, S., Kalpathy-Cramer, J., Farahani, K., Kirby, J., Burren, Y., Porz, N., Slotboom, J., Wiest, R., Lanczi, L., Gerstner, E., Weber, M.-A., Arbel, T., Avants, B. B., Ayache, N., Buendia, P., Collins, D. L., Cordier, N., et al. (2015). The multimodal brain tumor image segmentation benchmark (BRATS). IEEE Transactions on Medical Imaging, 34(10), 1993–2024. https://doi.org/10.1109/TMI.2014.2377694
Milletari, F., Navab, N., & Ahmadi, S. A. (2016). V-Net: Fully convolutional neural networks for volumetric medical image segmentation. In 2016 Fourth International Conference on 3D Vision (3DV) (pp. 565–571). IEEE. https://doi.org/10.1109/3DV.2016.79
Minderer, M., Djolonga, J., Romijnders, R., Hubis, F., Zhai, X., Houlsby, N., Tran, D., & Lucic, M. (2021). Revisiting the calibration of modern neural networks (Version 2). arXiv. https://doi.org/10.48550/arXiv.2106.07998
Oktay, O., Schlemper, J., Folgoc, L. L., Lee, M., Heinrich, M., Misawa, K., Mori, K., McDonagh, S., Hammerla, N. Y., Kainz, B., Glocker, B., & Rueckert, D. (2018). Attention U-Net: Learning where to look for the pancreas (Version 3). arXiv. https://doi.org/10.48550/arXiv.1804.03999
Ronneberger, O., Fischer, P., & Brox, T. (2015). U-Net: Convolutional networks for biomedical image segmentation. In Lecture Notes in Computer Science (pp. 234–241). Springer International Publishing, Cham, Switzerland. https://doi.org/10.1007/978-3-319-24574-4_28
Salehi, S. S. M., Erdogmus, D., & Gholipour, A. (2017). Tversky loss function for image segmentation using 3D fully convolutional deep networks. In Lecture Notes in Computer Science (pp. 379–387). Springer International Publishing, Cham, Switzerland. https://doi.org/10.1007/978-3-319-67389-9_44
Singh, A., Bay, A., Sengupta, B., & Mirabile, A. (2021). On deep neural network calibration by regularization and its impact on refinement (Version 3). arXiv. https://doi.org/10.48550/arXiv.2106.09385
Wang, C. (2023). Calibration in deep learning: A survey of the state-of-the-art (Version 4). arXiv. https://doi.org/10.48550/arXiv.2308.01222
Wang, D., Gong, B., & Wang, L. (2023). On calibrating semantic segmentation models: Analyses and an algorithm. In 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (pp. 23652–23662). IEEE. https://doi.org/10.1109/CVPR52729.2023.02265