Abstract
Automating cardiovascular disease diagnosis via medical image segmentation could save many lives across the world by bringing more cost-efficient technologies to developing countries. While neural networks currently perform well on many tasks, they can often fail when deployed to other domains, such as patients with different diseases. Therefore the M&Ms-2 challenge was created to help aid this effort to develop a neural network robust to domain shift in cardiac imaging. We use deformable Bayesian convolutional networks (DBCNs) [5] inserted into the nnUNet [6] framework to approach this problem and provide a solution to this challenge. We then explore the effects of both training time and network size on generalizability for both 2D long-axis and 3D short axis cardiac MRIs and find that the optimum training and network configuration is dependent both on the dataset size and task. We then enter the final test set of the competition to achieve competitive results .
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Campello, V.M., et al.: Multi-centre, multi-vendor and multi-disease cardiac segmentation: the m&ms challenge. IEEE Trans. Med. Imaging 40, 1 (2021). https://doi.org/10.1109/TMI.2021.3090082
Chen, C., et al.: Deep learning for cardiac image segmentation: a review. Front. Cardiovasc. Med. 7, 25 (2020)
D’Amour, A., et al.: Underspecification presents challenges for credibility in modern machine learning (2020)
Full, P.M., Isensee, F., Jäger, P.F., Maier-Hein, K.: Studying robustness of semantic segmentation under domain shift in cardiac MRI (2020)
Fulton, M., Heckman, C., Rentschler, M.: Deformable bayesian convolutional networks enable domain shift robustness in small training sets (2021). in review
Isensee, F., Jaeger, P.F., Kohl, S.A., Petersen, J., Maier-Hein, K.H.: NNU-NET: a self-configuring method for deep learning-based biomedical image segmentation. Nature Methods 18(2), 203–211 (2021)
Ma, J.: Histogram matching augmentation for domain adaptation with application to multi-centre, multi-vendor and multi-disease cardiac image segmentation (2020)
Petitjean, C., et al.: Right ventricle segmentation from cardiac MRI: a collation study. Med. Image Anal. 19(1), 187–202 (2015)
Shridhar, K., Laumann, F., Liwicki, M.: A comprehensive guide to bayesian convolutional neural network with variational inference (2019)
WHO: Cardiovascular diseases (CVD). https://www.who.int/news-room/fact-sheets/detail/cardiovascular-diseases-(cvds). Accessed 26 July 2021
Zhang, Y., et al.: Semi-supervised cardiac image segmentation via label propagation and style transfer (2020)
Zhou, K., Liu, Z., Qiao, Y., Xiang, T., Loy, C.C.: Domain generalization: a survey. arXiv preprint arXiv:2103.02503 (2021)
Zhu, X., Hu, H., Lin, S., Dai, J.: Deformable convnets v2: more deformable, better results (2018)
Acknowledgements
This research was supported in part by the National Science Foundation (NSF) (Grant No. 1849357). This work also used the Extreme Science and Engineering Discovery Environment (XSEDE), supported by National Science Foundation grant number ACI-1548562.
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
Fulton, M.J., Heckman, C.R., Rentschler, M.E. (2022). Deformable Bayesian Convolutional Networks for Disease-Robust Cardiac MRI Segmentation. In: Puyol Antón, E., et al. Statistical Atlases and Computational Models of the Heart. Multi-Disease, Multi-View, and Multi-Center Right Ventricular Segmentation in Cardiac MRI Challenge. STACOM 2021. Lecture Notes in Computer Science(), vol 13131. Springer, Cham. https://doi.org/10.1007/978-3-030-93722-5_32
Download citation
DOI: https://doi.org/10.1007/978-3-030-93722-5_32
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-93721-8
Online ISBN: 978-3-030-93722-5
eBook Packages: Computer ScienceComputer Science (R0)
