Generalised Dice Overlap as a Deep Learning Loss Function for Highly Unbalanced Segmentations
Open accessAbstract
Deep-learning has proved in recent years to be a powerful tool for image analysis and is now widely used to segment both 2D and 3D medical images. Deep-learning segmentation frameworks rely not only on the choice of network architecture but also on the choice of loss function. When the segmentation process targets rare observations, a severe class imbalance is likely to occur between candidate labels, thus resulting in sub-optimal performance. In order to mitigate this issue, strategies such as the weighted cross-entropy function, the sensitivity function or the Dice loss function, have been proposed. In this work, we investigate the behavior of these loss functions and their sensitivity to learning rate tuning in the presence of different rates of label imbalance across 2D and 3D segmentation tasks. We also propose to use the class re-balancing properties of the Generalized Dice overlap, a known metric for segmentation assessment, as a robust and accurate deep-learning loss function for unbalanced tasks.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
U-Net: Convolutional Networks for Biomedical Image Segmentation
Ronneberger, O., Fischer, P., & Brox, T. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-24574-4_28 · Cited by 99,404 · see citing works
Efficient multi-scale 3D CNN with fully connected CRF for accurate brain lesion segmentation
Kamnitsas, K., Ledig, C., Newcombe, V. F. J., Simpson, J. P., Kane, A. D., Menon, D. K., Rueckert, D., & Glocker, B. (2017) · Medical Image Analysis · doi:10.1016/j.media.2016.10.004 · Cited by 3,498 · cited 1× · see citing works
“3D networks: The DeepMedic architecture [4] and the HighResNet network [6] were used in the 3D context.”
MethodBrain tumor segmentation with Deep Neural Networks
Havaei, M., Davy, A., Warde-Farley, D., Biard, A., Courville, A., Bengio, Y., Pal, C., Jodoin, P.-M., & Larochelle, H. (2017) · Medical Image Analysis · doi:10.1016/j.media.2016.05.004 · Cited by 3,278 · see citing works
Generalized Overlap Measures for Evaluation and Validation in Medical Image Analysis
Crum, W. R., Camara, O., & Hill, D. L. G. (2006) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2006.880587 · Cited by 776 · see citing works
Flagged first, then by citations
Research works printed but not linked: 6
- [1]Brosch T Yoo Y Tang LYW Li DKB Traboulsee A Tam R Deep convolutional encoder networks for multiple sclerosis lesion segmentation Navab N Hornegger J Wells WM Frangi AF MICCAI 2015. LNCS Springer Cham 2015 9351 3 11 10.1007/978-3-319-24574-4_1 · DOI
- [5]Lai M Deep learning for medical image segmentation arXiv:1505.02000 2015 · arXiv
- [6]Li W Wang G Fidon L Ourselin S Cardoso MJ Vercauteren T On the compactness, efficiency, and representation of 3D convolutional networks: brain parcellation as a pretext task Niethammer M Styner M Aylward S Zhu H Oguz I Yap P-T Shen D IPMI 2017. LNCS Springer Cham 2017 10265 348 360 10.1007/978-3-319-59050-9_28 · DOI
- [7]Menze BH The multimodal brain tumor image segmentation benchmark (BRATS) IEEE TMI 2015 34 10 1993 2024 PMC4833122 25494501
- [8]Milletari F Navab N Ahmadi SA V-Net: fully convolutional neural networks for volumetric medical image segmentation 2016 Fourth International Conference on 3D Vision (3DV) IEEE 2016 10 565 571
- [10]Zheng Y Liu D Georgescu B Nguyen H Comaniciu D 3D deep learning for efficient and robust landmark detection in volumetric data Navab N Hornegger J Wells WM Frangi AF MICCAI 2015. LNCS Springer Cham 2015 9349 565 572 10.1007/978-3-319-24553-9_69 · DOI
Cited by
Sorted by citations
Loss odyssey in medical image segmentation
Ma, J., Chen, J., Ng, M., Huang, R., Li, Y., Li, C., Yang, X., & Martel, A. L. (2021) · Medical Image Analysis · doi:10.1016/j.media.2021.102035 · Cited by 662 · see citing works
Mazurowski, M. A., Buda, M., Saha, A., & Bashir, M. R. (2019) · Journal of Magnetic Resonance Imaging · doi:10.1002/jmri.26534 · Cited by 554 · see citing works
NiftyNet: a deep-learning platform for medical imaging
Gibson, E., Li, W., Sudre, C., Fidon, L., Shakir, D. I., Wang, G., Eaton-Rosen, Z., Gray, R., Doel, T., Hu, Y., et al. (2018) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2018.01.025 · Cited by 552 · see citing works
Self-citationConfidence calibration and predictive uncertainty estimation for deep medical image segmentation
Mehrtash, A., Wells, W. M., Tempany, C. M., Abolmaesumi, P., & Kapur, T. (2020) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2020.3006437 · Cited by 428 · see citing works
Federated learning enables big data for rare cancer boundary detection
Pati, S., Baid, U., Edwards, B., Sheller, M., Wang, S.-H., Reina, G. A., Foley, P., Gruzdev, A., Karkada, D., Davatzikos, C., et al. (2022) · Nature Communications · doi:10.1038/s41467-022-33407-5 · Cited by 394 · see citing works
Towards global flood mapping onboard low cost satellites with machine learning
Mateo-Garcia, G., Veitch-Michaelis, J., Smith, L., Oprea, S. V., Schumann, G., Gal, Y., Baydin, A. G., & Backes, D. (2021) · Scientific Reports · doi:10.1038/s41598-021-86650-z · Cited by 202 · see citing works
Celard, P., Iglesias, E. L., Sorribes-Fdez, J. M., Romero, R., Vieira, A. S., & Borrajo, L. (2023) · Neural Computing and Applications · doi:10.1007/s00521-022-07953-4 · Cited by 188 · see citing works
Morpheus: A Deep Learning Framework for the Pixel-level Analysis of Astronomical Image Data
Hausen, R., & Robertson, B. E. (2020) · The Astrophysical Journal Supplement Series · doi:10.3847/1538-4365/ab8868 · Cited by 110 · see citing works
“A common approach to handle unbalanced segmentations is to employ a Dice loss function to supplement the entropy loss function (e.g., Milletari et al. 2016; Sudre et al. 2017).”
MethodFakeLocator: Robust Localization of GAN-Based Face Manipulations
Huang, Y., Juefei-Xu, F., Guo, Q., Liu, Y., & Pu, G. (2022) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2022.3141262 · Cited by 102 · see citing works
Pati, S., Baid, U., Edwards, B., Sheller, M. J., Foley, P., Anthony Reina, G., Thakur, S., Sako, C., Bilello, M., Davatzikos, C., et al. (2022) · Physics in Medicine & Biology · doi:10.1088/1361-6560/ac9449 · Cited by 51 · see citing works
Mapping Roads in the Brazilian Amazon with Artificial Intelligence and Sentinel-2
Botelho, J., Costa, S. C. P., Ribeiro, J. G., & Souza, C. M. (2022) · Remote Sensing · doi:10.3390/rs14153625 · Cited by 43 · see citing works
Lee, W.-K., Wu, C.-C., Lee, C.-C., Lu, C.-F., Yang, H.-C., Huang, T.-H., Lin, C.-Y., Chung, W.-Y., Wang, P.-S., Wu, H.-M., et al. (2020) · Artificial Intelligence in Medicine · doi:10.1016/j.artmed.2020.101911 · Cited by 35 · see citing works
Joo, L., Shim, W. H., Suh, C. H., Lim, S. J., Heo, H., Kim, W. S., Hong, E., Lee, D., Sung, J., Lim, J.-S., et al. (2022) · PLOS ONE · doi:10.1371/journal.pone.0274562 · Cited by 35 · see citing works
Cho, J., Kim, Y. J., Sunwoo, L., Lee, G. P., Nguyen, T. Q., Cho, S. J., Baik, S. H., Bae, Y. J., Choi, B. S., Jung, C., et al. (2021) · Frontiers in Oncology · doi:10.3389/fonc.2021.739639 · Cited by 30 · see citing works
Deep attentive fusion network for flood detection on uni-temporal Sentinel-1 data
Yadav, R., Nascetti, A., & Ban, Y. (2022) · Frontiers in Remote Sensing · doi:10.3389/frsen.2022.1060144 · Cited by 21 · see citing works
clDice - a Novel Topology-Preserving Loss Function for Tubular Structure Segmentation
Shit, S., Paetzold, J. C., Sekuboyina, A., Ezhov, I., Unger, A., Zhylka, A., Pluim, J. P. W., Bauer, U., & Menze, B. H. (2021) · 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr46437.2021.01629 · Cited by 60 · see citing works
TARig: Adaptive template-aware neural rigging for humanoid characters
Ma, J., & Zhang, D. (2023) · doi:10.2139/ssrn.4326447 · Cited by 17
Efficient Model Monitoring for Quality Control in Cardiac Image Segmentation
Galati, F., & Zuluaga, M. A. (2021) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-78710-3_11 · Cited by 15 · see citing works
Ego-Exo4D: Understanding Skilled Human Activity from First- and Third-Person Perspectives
Grauman, K., Westbury, A., Torresani, L., Kitani, K., Malik, J., Afouras, T., Ashutosh, K., Baiyya, V., Bansal, S., Boote, B., et al. (2025) · International Journal of Computer Vision · doi:10.1007/s11263-025-02557-6 · Cited by 3 · see citing works
Skin Tokens: A Learned Compact Representation for Unified Autoregressive Rigging
Zhang, J.-P., Pu, C.-F., Guo, M.-H., Cao, Y.-P., & Hu, S.-M. (2026) · arXiv · doi:10.48550/arxiv.2602.04805
- Authors
- Sudre, C. H., Li, W., Vercauteren, T., Ourselin, S., & Jorge Cardoso, M.
- Affiliations
- UCL Queen Square Institute of Neurology (GB)
- University College London (GB)
- Corresponding author
- Sudre, Carole H.
- Venue
- Lecture Notes in Computer Science
- Publisher
- Springer International Publishing
- Published
- 2017
- Volume
- 2017
- Pages
- 240-248
- ISSN
- 0302-9743 (ISSN-L)
- 1611-3349
- PMC
- PMC7610921
- OpenAlex
- W2734349601
- Cited by
- 2,829
- References
- 4
- Citations by year
- 2026: 348 · 2025: 425 · 2024: 461 · 2023: 416 · 2022: 374 · 2021: 369 · 2020: 241 · 2019: 154 · 2018: 39 · 2017: 2
- FWCI
- 34.60
- Type
- conference paper
- Language
- en
- Open access
- green
- www.ncbi.nlm.nih.gov
- Locations
- Lecture notes in computer science (publishedVersion)
- PubMed (publishedVersion)
- PubMed Central (acceptedVersion)
- arXiv (Cornell University) (submittedVersion)
- UCL Discovery (University College London) (submittedVersion)
- Research Portal (King's College London) (submittedVersion, other-oa)
- arXiv (Cornell University)
- Funders
- Wellcome Trust (EP/H046410/1, WT101957, EP/J020990/1, 101957)
- National Institute for Health and Care Research (BRC10269, EP/J020990/1)
- University College London (EP/J020990/1, EP/H046410/1, EP/K005278)
- University College London Hospitals NHS Foundation Trust
- Medical Research Council (MR/J01107X/, FP7-ICT-2011-9-601055, MR/J01107X/1)
- Engineering and Physical Sciences Research Council (EP/H046410, EP/K005278, H046410, MR/J01107X/1, EP/M020533/1, EP/H046410/1, EP/J020990/1, FP7-ICT-2011-9-601055, J020990, WT101957)
- Science and Technology Facilities Council
- UCLH Biomedical Research Centre
- FP7 Information and Communication Technologies (FP7‐ICT‐2011‐9‐601055)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Medical Image Segmentation Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Keywords
- loss functions, class imbalance, medical image segmentation, 3D segmentation, learning rate sensitivity, weighted cross-entropy, sensitivity function, Dice loss
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
