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

21 May 2021  ·  Shumeng Li, Ziyuan Zhao, Kaixin Xu, Zeng Zeng, Cuntai Guan ·

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.

PDF Abstract


  Add Datasets introduced or used in this paper
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Left Atrium Segmentation Atrial Segmentation Challenge HCR-MT 95HD 6.93 # 2
ASD 2.18 # 2
Dice Score 90.04 # 1
Jaccard 81.98 # 1


No methods listed for this paper. Add relevant methods here