Abstract
Segmentation of cardiac structures in magnetic resonance imaging is essential to the diagnosis of many cardiovascular diseases. However, sometimes it is challenging to accurately define the right ventricle (RV) structure due to the complex texture. It requires the collaboration of both short-axis (SA) images and long-axis (LA) images. Current deep learning methods trained with single-view data neglect the spatial relations between SA and LA images and could fail at the basal and apex plane of the RV. In order to properly handle the geometrical relations, we proposed a consistency based co-training method that involves an extra penalty for confusing images at the basal and apex plane. At test phase, the anatomical relations are also used for post-processing which further improves the robustness of segmentation. Besides, we incorporate the transformer network in U-Net architecture to enhance its ability to model long-range dependency. We evaluated the proposed method in the M&Ms-2 challenge and obtained promising performance for the segmentation of both SA and LA images.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
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
Carion, N., Massa, F., Synnaeve, G., Usunier, N., Kirillov, A., Zagoruyko, S.: End-to-end object detection with transformers (2020)
Chen, C., Biffi, C., Tarroni, G., Petersen, S., Bai, W., Rueckert, D.: Learning shape priors for robust cardiac MR segmentation from multi-view images. Med. Image Comput. Comput. Assist. Interv. MICCAI 2019, 523–531 (2019)
Chen, J., et al.: Transunet: transformers make strong encoders for medical image segmentation (2021)
Gehring, J., Auli, M., Grangier, D., Yarats, D., Dauphin, Y.N.: Convolutional sequence to sequence learning (2017)
Pop, M., et al. (eds.): STACOM 2017. LNCS, vol. 10663. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-75541-0
Khan, S., Naseer, M., Hayat, M., Zamir, S.W., Khan, F.S., Shah, M.: Transformers in vision: a survey (2021)
Ronneberger, O., Fischer, P., Brox, T.: U-net: convolutional networks for biomedical image segmentation (2015)
Schlemper, J., et al.: Attention gated networks: learning to leverage salient regions in medical images (2019)
Yu, Q., Xie, L., Wang, Y., Zhou, Y., Fishman, E.K., Yuille, A.L.: Recurrent saliency transformation network: incorporating multi-stage visual cues for small organ segmentation (2018)
Zhao, H., Shi, J., Qi, X., Wang, X., Jia, J.: Pyramid scene parsing network (2017)
Zheng, S., et al.: Rethinking semantic segmentation from a sequence-to-sequence perspective with transformers (2021)
Zhou, Z., Siddiquee, M.M.R., Tajbakhsh, N., Liang, J.: Unet++: a nested u-net architecture for medical image segmentation (2018)
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
Gao, Z., Zhuang, X. (2022). Consistency Based Co-segmentation for Multi-view Cardiac MRI Using Vision Transformer. 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_33
Download citation
DOI: https://doi.org/10.1007/978-3-030-93722-5_33
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)