Abstract
Albeit the Dice loss is one of the dominant loss functions in medical image segmentation, most research omits a closer look at its derivative, i.e. the real motor of the optimization when using gradient descent. In this paper, we highlight the peculiar action of the Dice loss in the presence of missing or empty labels. First, we formulate a theoretical basis that gives a general description of the Dice loss and its derivative. It turns out that the choice of the reduction dimensions \(\varPhi \) and the smoothing term \(\epsilon \) is non-trivial and greatly influences its behavior. We find and propose heuristic combinations of \(\varPhi \) and \(\epsilon \) that work in a segmentation setting with either missing or empty labels. Second, we empirically validate these findings in a binary and multiclass segmentation setting using two publicly available datasets. We confirm that the choice of \(\varPhi \) and \(\epsilon \) is indeed pivotal. With \(\varPhi \) chosen such that the reductions happen over a single batch (and class) element and with a negligible \(\epsilon \), the Dice loss deals with missing labels naturally and performs similarly compared to recent adaptations specific for missing labels. With \(\varPhi \) chosen such that the reductions happen over multiple batch elements or with a heuristic value for \(\epsilon \), the Dice loss handles empty labels correctly. We believe that this work highlights some essential perspectives and hope that it encourages researchers to better describe their exact implementation of the Dice loss in future work.
S. Tilborghs and J. Bertels—Contributed equally to this work.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Bakas, S., et al.: Advancing The Cancer Genome Atlas glioma MRI collections with expert segmentation labels and radiomic features. Sci. Data 4(1), 170117 (2017). https://doi.org/10.1038/sdata.2017.117, http://www.nature.com/articles/sdata2017117
Bakas, S., Reyes, M., Jakab, A., Bauer, S., Rempfler, M., et al.: Identifying the best machine learning algorithms for brain tumor segmentation, progression assessment, and overall survival prediction in the BRATS challenge. arXiv, November 2018. http://arxiv.org/abs/1811.02629
Bernard, O., et al.: Deep learning techniques for automatic MRI cardiac multi-structures segmentation and diagnosis: is the problem solved? IEEE Trans. Med. Imaging 37(11), 2514–2525 (2018). https://doi.org/10.1109/TMI.2018.2837502
Bertels, J., Robben, D., Vandermeulen, D., Suetens, P.: Theoretical analysis and experimental validation of volume bias of soft Dice optimized segmentation maps in the context of inherent uncertainty. Med. Image Anal. 67, 101833 (2021). https://doi.org/10.1016/j.media.2020.101833, https://linkinghub.elsevier.com/retrieve/pii/S1361841520301973
Drozdzal, M., Vorontsov, E., Chartrand, G., Kadoury, S., Pal, C.: The importance of skip connections in biomedical image segmentation. In: Carneiro, G., et al. (eds.) LABELS/DLMIA -2016. LNCS, vol. 10008, pp. 179–187. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-46976-8_19
Eelbode, T., et al.: Optimization for medical image segmentation: theory and practice when evaluating with dice score or Jaccard index. IEEE Trans. Med. Imaging 39(11), 3679–3690 (2020). https://doi.org/10.1109/TMI.2020.3002417, https://ieeexplore.ieee.org/document/9116807/
Fidon, L., et al.: Label-set loss functions for partial supervision: application to fetal brain 3D MRI parcellation. In: de Bruijne, M., et al. (eds.) MICCAI 2021. LNCS, vol. 12902, pp. 647–657. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-87196-3_60
Goodfellow, I., Bengio, Y., Courville, A.: Deep Learning. MIT Press, Cambridge (2016). http://www.deeplearningbook.org
Isensee, F., Kickingereder, P., Wick, W., Bendszus, M., Maier-Hein, K.H.: No New-Net. In: Crimi, A., Bakas, S., Kuijf, H., Keyvan, F., Reyes, M., van Walsum, T. (eds.) BrainLes 2018. LNCS, vol. 11384, pp. 234–244. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-11726-9_21
Jadon, S.: A survey of loss functions for semantic segmentation. In: 2020 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology, CIBCB 2020 (2020). https://doi.org/10.1109/CIBCB48159.2020.9277638
Kodym, O., Španěl, M., Herout, A.: Segmentation of head and neck organs at risk using CNN with batch dice loss. In: Brox, T., Bruhn, A., Fritz, M. (eds.) GCPR 2018. LNCS, vol. 11269, pp. 105–114. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-12939-2_8
Menze, B.H., et al.: The multimodal brain tumor image segmentation benchmark (BRATS). IEEE Trans. Med. Imaging 34(10), 1993–2024 (2015). https://doi.org/10.1109/TMI.2014.2377694, http://ieeexplore.ieee.org/document/6975210/
Milletari, F., Navab, N., Ahmadi, S.A.: V-Net: fully convolutional neural networks for volumetric medical image segmentation. In: Proceedings - 2016 4th International Conference on 3D Vision, 3DV 2016, pp. 565–571 (2016). https://doi.org/10.1109/3DV.2016.79
Nowozin, S.: Optimal decisions from probabilistic models: the intersection-over-union case. In: 2014 IEEE Conference on Computer Vision and Pattern Recognition, pp. 548–555. IEEE, June 2014. https://doi.org/10.1109/CVPR.2014.77, http://ieeexplore.ieee.org/document/6909471/
Shi, G., Xiao, L., Chen, Y., Zhou, S.K.: Marginal loss and exclusion loss for partially supervised multi-organ segmentation. Med. Image Anal. 70, 101979 (2021). https://doi.org/10.1016/j.media.2021.101979
Sudre, C.H., Li, W., Vercauteren, T., Ourselin, S., Jorge Cardoso, M.: Generalised dice overlap as a deep learning loss function for highly unbalanced segmentations. In: Cardoso, M.J., et al. (eds.) DLMIA/ML-CDS -2017. LNCS, vol. 10553, pp. 240–248. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-67558-9_28
Tarlow, D., Adams, R.P.: Revisiting uncertainty in graph cut solutions. In: 2012 IEEE Conference on Computer Vision and Pattern Recognition, pp. 2440–2447. IEEE, June 2012. https://doi.org/10.1109/CVPR.2012.6247958, http://ieeexplore.ieee.org/document/6247958/
Yeung, M., Sala, E., Schönlieb, C.B., Rundo, L.: Unified focal loss: generalising dice and cross entropy-based losses to handle class imbalanced medical image segmentation. Computerized Med. Imaging Graph. 95, 102026 (2021, 2022). https://doi.org/10.1016/j.compmedimag.2021.102026
Acknowledgement
This research received funding from the Flemish Government under the “Onderzoeksprogramma Artificiële intelligentie (AI) Vlaanderen” programme and is also partially funded by KU Leuven Internal Funds C24/18/047 (F. Maes).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Tilborghs, S., Bertels, J., Robben, D., Vandermeulen, D., Maes, F. (2022). The Dice Loss in the Context of Missing or Empty Labels: Introducing \(\varPhi \) and \(\epsilon \). In: Wang, L., Dou, Q., Fletcher, P.T., Speidel, S., Li, S. (eds) Medical Image Computing and Computer Assisted Intervention – MICCAI 2022. MICCAI 2022. Lecture Notes in Computer Science, vol 13435. Springer, Cham. https://doi.org/10.1007/978-3-031-16443-9_51
Download citation
DOI: https://doi.org/10.1007/978-3-031-16443-9_51
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-16442-2
Online ISBN: 978-3-031-16443-9
eBook Packages: Computer ScienceComputer Science (R0)