.png){kind=icon}
返回
Collaborative Learning for Annotation-Efficient Volumetric MR Image Segmentation
.png){kind=icon}
PRE
AI
DOI：10.1002/jmri.29194.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Background: Deep learning has presented great potential in accurate MR image segmentation when enough labeled data are provided for network optimization. However, manually annotating three-dimensional (3D) MR images is tedious and time-consuming, requiring experts with rich domain knowledge and experience.Purpose: To build a deep learning method exploring sparse annotations, namely only a single two-dimensional slice label for each 3D training MR image.Study Type: Retrospective.Population: Three-dimensional MR images of 150 subjects from two publicly available datasets were included. Among them, 50 (1377 image slices) are for prostate segmentation. The other 100 (8800 image slices) are for left atrium segmentation. Five-fold cross-validation experiments were carried out utilizing the first dataset. For the second dataset, 80 subjects were used for training and 20 were used for testing.Field Strength/Sequence: 1.5 T and 3.0 T; axial T2-weighted and late gadolinium-enhanced, 3D respiratory navigated, inversion recovery prepared gradient echo pulse sequence.Assessment: A collaborative learning method by integrating the strengths of semi-supervised and self-supervised learning schemes was developed. The method was trained using labeled central slices and unlabeled noncentral slices. Segmentation performance on testing set was reported quantitatively and qualitatively.Statistical Tests: Quantitative evaluation metrics including boundary intersection-over-union (B-IoU), Dice similarity coefficient, average symmetric surface distance, and relative absolute volume difference were calculated. Paired t test was performed, and P < 0.05 was considered statistically significant.Results: Compared to fully supervised training with only the labeled central slice, mean teacher, uncertainty-aware mean teacher, deep co-training, interpolation consistency training (ICT), and ambiguity-consensus mean teacher, the proposed method achieved a substantial improvement in segmentation accuracy, increasing the mean B-IoU significantly by more than 10.0% for prostate segmentation (proposed method B-IoU: 70.3% +/- 7.6% vs. ICT B-IoU: 60.3% +/- 11.2%) and by more than 6.0% for left atrium segmentation (proposed method B-IoU: 66.1% +/- 6.8% vs. ICT B-IoU: 60.1% +/- 7.1%).Data Conclusions: A collaborative learning method trained using sparse annotations can segment prostate and left atrium with high accuracy.
Keyword:
volumetric MR image segmentation
sparse annotations
pseudo labeling
semi-supervised learning
self-supervised learning
期刊
IF:
3.5
论文数:
10.0K
被引数:
2.0W
机构
引用论文
Understanding the accumulation features of POPs in squid from the offshore waters of southeast Korea
PREAI
分享收藏A Coarse-to-Fine Deformable Transformation Framework for Unsupervised Multi-Contrast MR Image Registration with Dual Consistency Constraint具有双重一致性约束的无监督多对比度MR图像配准的粗到细可变形变换框架
OAAI
OAAI分享收藏U-Net: deep learning for cell counting, detection, and morphometryU-net: 用于细胞计数、检测和形态测量的深度学习.png)
NATURE METHODS
NATURE METHODSIF32.1
PREAI
分享收藏Deep Learning Whole-Gland and Zonal Prostate Segmentation on a Public MRI Dataset在公共MRI数据集上深度学习全腺和带状前列腺分割
PREAI
分享收藏Annotation-efficient deep learning for automatic medical image segmentationNATURE COMMUNICATIONSOAAI
NATURE COMMUNICATIONSIF15.7
OAAI分享收藏Interpolation consistency training for semi-supervised learning半监督学习的插值一致性训练NEURAL NETWORKSOAAI
NEURAL NETWORKSIF6.3
OAAI分享收藏A global benchmark of algorithms for segmenting the left atrium from late gadolinium-enhanced cardiac magnetic resonance imaging从晚期g增强心脏磁共振成像中分割左心房的算法的全球基准MEDICAL IMAGE ANALYSISOAAI
MEDICAL IMAGE ANALYSISIF11.8
OAAI分享收藏Quantifying the generalization error in deep learning in terms of data distribution and neural network smoothness根据数据分布和神经网络平滑度量化深度学习中的泛化误差NEURAL NETWORKSOAAI
NEURAL NETWORKSIF6.3
OAAI分享收藏
查看更多内容