Abstract
As the third-place winning method for the MIDOG mitosis detection challenge, we created a cascade algorithm consisting of a Mask-RCNN detector, followed by a classification ensemble consisting of ResNet50 and DenseNet201 to refine detected mitotic candidates. The MIDOG training data consists of 200 frames originating from four scanners, three of which are annotated for mitotic instances with centroid annotations. Our main algorithmic choices are as follows: first, to enhance the generalizability of our detector and classification networks, we use a state-of-the-art Residual Cycle-GAN to transform each scanner domain to every other scanner domain. During training, we then randomly load, for each image, one of the four domains. In this way, our networks can learn from the fourth non-annotated scanner domain even if we don’t have annotations for it. Second, for training the detector network, rather than using centroid-based fixed-size bounding boxes, we create mitosis-specific bounding boxes. We do this by manually annotating a small selection of mitoses, training a Mask-RCNN on this small dataset, and applying it to the rest of the data to obtain full annotations. We trained the follow-up classification ensemble using only the challenge-provided positive and hard-negative examples. On the preliminary and final test set, the algorithm scores an F1 score of 0.7578 and 0.7361, respectively, putting us as the preliminary second-place and final third-place team on the leaderboard.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Aubreville, M., et al.: MItosis DOmain Generalization challenge (MIDOG). In: 24th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) (2021). https://doi.org/10.5281/zenodo.4573978
Aubreville, M., et al.: Quantifying the scanner-induced domain gap in mitosis detection. In: Medical Imaging with Deep Learning (MIDL) (2021)
de Bel, T., et al.: Residual cyclegan for robust domain transformation of histopathological tissue slides. Med. Image Anal. 70, 102004 (2021)
Bertram, C.A., et al.: Are pathologist-defined labels reproducible? Comparison of the TUPAC16 mitotic figure dataset with an alternative set of labels. In: Cardoso, J., et al. (eds.) IMIMIC/MIL3ID/LABELS -2020. LNCS, vol. 12446, pp. 204–213. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-61166-8_22
Ciga, O., Xu, T., Martel, A.L.: Self supervised contrastive learning for digital histopathology. Mach. Learn. Appl. 7, 100198 (2021)
Faryna, K., van der Laak, J., Litjens, G.: Tailoring automated data augmentation to H&E-stained histopathology. In: Medical Imaging with Deep Learning (2021)
He, K., et al.: Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770–778 (2016)
He, K., et al.: Mask R-CNN. In: Proceedings of the IEEE ICCV, pp. 2961–2969 (2017)
Huang, G., et al.: Densely connected convolutional networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 4700–4708 (2017)
Kausar, T., et al.: SmallMitosis: small size mitotic cells detection in breast histopathology images. IEEE Access 9, 905–922 (2020)
MITOS14 Challenge (2014). https://mitos-atypia-14.grand-challenge.org/
Ren, S., et al.: Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks (2016). arXiv: 1506.01497 [cs.CV]
Roux, L., et al.: Mitosis detection in breast cancer histological images an ICPR 2012 contest. J. Pathol. Inform. 4 (2013)
Veta, M., et al.: Assessment of algorithms for mitosis detection in breast cancer histopathology images. Med. Image Anal. 20(1), 237–248 (2015)
Veta, M., et al.: Predicting breast tumor proliferation from whole-slide images: the TUPAC16 challenge. Med. Image Anal. 54, 111–121 (2019)
Wilm, F., Breininger, K., Aubreville, M.: Domain adversarial RetinaNet as a reference algorithm for the MItosis DOmain Generalization (MIDOG) challenge. In: Biomedical Image Registration, Domain Generalisation and Out-of-Distribution Analysis, MICCAI 2021 Challenges L2R, MIDOG and MOOD (2021)
Wilm, F., et al.: Influence of inter-annotator variability on automatic mitotic figure assessment. In: Palm, C., Deserno, T.M., Handels, H., Maier, A., Maier-Hein, K., Tolxdorff, T. (eds.) Bildverarbeitung für die Medizin 2021. I, pp. 241–246. Springer, Wiesbaden (2021). https://doi.org/10.1007/978-3-658-33198-6_56
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 Springer Nature Switzerland AG
About this paper
Cite this paper
Fick, R.H.J., Moshayedi, A., Roy, G., Dedieu, J., Petit, S., Hadj, S.B. (2022). Domain-Specific Cycle-GAN Augmentation Improves Domain Generalizability for Mitosis Detection. In: Aubreville, M., Zimmerer, D., Heinrich, M. (eds) Biomedical Image Registration, Domain Generalisation and Out-of-Distribution Analysis. MICCAI 2021. Lecture Notes in Computer Science(), vol 13166. Springer, Cham. https://doi.org/10.1007/978-3-030-97281-3_5
Download citation
DOI: https://doi.org/10.1007/978-3-030-97281-3_5
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-97280-6
Online ISBN: 978-3-030-97281-3
eBook Packages: Computer ScienceComputer Science (R0)