Challenge report
Marc Aubreville, Nikolas Stathonikos, Christof A. Bertram, Robert Klopfleisch, Natalie ter Hoeve, Francesco Ciompi, Frauke Wilm, Christian Marzahl, Taryn A. Donovan, Andreas Maier, Jack Breen, Nishant Ravikumar, Youjin Chung, Jinah Park, Ramin Nateghi, Fattaneh Pourakpour, Rutger H. J. Fick, Saima Ben Hadj, Mostafa Jahanifar, Nasir Rajpoot, Jakob Dexl, Thomas Wittenberg, Satoshi Kondo, Maxime W. Lafarge, Viktor H. Koelzer, Jingtang Liang, Yubo Wang, Xi Long, Jingxin Liu, Salar Razavi, April Khademi, Sen Yang, Xiyue Wang, Mitko Veta, Katharina Breininger: Mitosis domain generalization in histopathology images — The MIDOG challenge. Medical Image Analysis 84 (2023), 102699. DOI: 10.1016/j.media.2022.102699
- Journal version: https://www.sciencedirect.com/science/article/abs/pii/S1361841522003279
- Preprint: arXiv:2204.03742
Structured challenge description
Aubreville, M., Bertram, C., Veta, M., Klopfleisch, R., Stathonikos, N., Breininger, K., ter Hoeve, N., Ciompi, F., Maier, A.: Mitosis domain generalization challenge. 24th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2021), 2021. DOI: 10.5281/zenodo.4573978
Dataset paper
A comprehensive multi-domain dataset for mitotic figure detection. Scientific Data, 2023.
Workshop proceedings (MICCAI 2021)
All participants of the MIDOG challenge may submit a short LNCS Springer paper (up to 4 pages) with a deadline of three weeks after the workshop, which will be subject to peer review. (grand-challenge.org)
The peer-reviewed papers of the reference approach and the participating teams were published in: Biomedical Image Registration, Domain Generalisation and Out-of-Distribution Analysis (MICCAI 2021 challenges L2R, MIDOG and MOOD). Lecture Notes in Computer Science, vol. 13166, Springer, Cham, 2022.
Reference approach
- Wilm, F., Marzahl, C., Breininger, K., Aubreville, M.: Domain adversarial retinanet as a reference algorithm for the midog challenge. pp. 5–13. Reference Docker container: https://github.com/DeepPathology/MIDOG_reference_docker
Participant papers
| Team | Paper | Pages |
|---|---|---|
| Leeds | Breen, J., Zucker, K., Orsi, N.M., Ravikumar, N.: Assessing domain adaptation techniques for mitosis detection in multi-scanner breast cancer histopathology images | 14–22 |
| CGV | Chung, Y., Cho, J., Park, J.: Domain-robust mitotic figure detection with style transfer | 23–31 |
| PixelPath-AI | Nateghi, R., Pourakpour, F.: Two-step domain adaptation for mitosis cell detection in histopathology images | 32–39 |
| Tribun Healthcare | Fick, R.H., Moshayedi, A., Roy, G., Dedieu, J., Petit, S., Hadj, S.B.: Domain-specific cycle-gan augmentation improves domain generalizability for mitosis detection | 40–47 |
| TIA Centre | Jahanifar, M., Shepard, A., Zamanitajeddin, N., Bashir, R.S., Bilal, M., Khurram, S.A., Minhas, F., Rajpoot, N.: Stain-robust mitotic figure detection for the Mitosis Domain Generalization Challenge | 48–52 |
| jdex | Dexl, J., Benz, M., Bruns, V., Kuritcy, P., Wittenberg, T.: MitoDet: Simple and robust mitosis detection | 53–57 |
| SK | Kondo, S.: Multi-source domain adaptation using gradient reversal layer for mitotic cell detection | 58–61 |
| ML | Lafarge, M., Koelzer, V.: Rotation invariance and extensive data augmentation: a strategy for the MItosis DOmain Generalization (MIDOG) Challenge | 62–67 |
| XidianU-OUC | Liang, J., Wang, C., Cheng, Y., Wang, Z., Wang, F., Huang, L., Yu, Z., Wang, Y.: Detecting mitosis against domain shift using a fused detector and deep ensemble classification model for MIDOG challenge | 68–72 |
| No.0 | Long, X., Cheng, Y., Mu, X., Liu, L., Liu, J.: Domain adaptive cascade R-CNN for MItosis DOmain Generalization (MIDOG) Challenge | 73–76 |
| IAMLAB | Razavi, S., Dambandkhameneh, F., Androutsos, D., Done, S., Khademi, A.: Cascade R-CNN for MIDOG challenge | 81–85 |
| AI medical | Yang, S., Luo, F., Zhang, J., Wang, X.: Sk-unet model with fourier domain for mitosis detection | 86–90 |