Hierarchical Consistency Regularized Mean Teacher for Semi-supervised 3D Left Atrium Segmentation

Hierarchical Consistency Regularized Mean Teacher for Semi-supervised 3D Left Atrium Segmentation

2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC)

10.1109/EMBC46164.2021.9629941

https://doi.org/10.1109/embc46164.2021.9629941

Shumeng Li, Ziyuan Zhao, Kaixin Xu, Zeng Zeng, Cuntai Guan

deep learning, deep convolutional neural networks, Semi supervised learning, Medical image segmentation

09 December 2021

Li, S., Zhao, Z., Xu, K., Zeng, Z., & Guan, C. (2021). Hierarchical Consistency Regularized Mean Teacher for Semi-supervised 3D Left Atrium Segmentation. 2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC). doi:10.1109/embc46164.2021.9629941

Deep learning has achieved promising segmentation performance on 3D left atrium MR images. However, annotations for segmentation tasks are expensive, costly and difficult to obtain. In this paper, we introduce a novel hierarchical consistency regularized mean teacher framework for 3D left atrium segmentation. In each iteration, the student model is optimized by multi-scale deep supervision and hierarchical consistency regularization, concurrently. Extensive experiments have shown that our method achieves competitive performance as compared with full annotation, outperforming other state-of-the-art semi-supervised segmentation methods.

Publisher Copyright

This research is supported by core funding from: Institute for Infocomm Research
Grant Reference no. :

© 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

https://oar.a-star.edu.sg/communities-collections/articles/18078

2694-0604
2375-7477

978-1-7281-1179-7
978-1-7281-1178-0
978-1-7281-1180-3

Institute for Infocomm Research