IEEE Transactions on Medical Imaging · 2024 · 35 citations · 32 references
ComputingBiomedical EngineeringComputer ScienceBiomedical ImagingComputer VisionMachine LearningAnnotationDeep LearningImage AnnotationSelf-supervised LearningSupervised LearningVolume SegmentationGround Truth LabelingLatent Space ExplorationOrgan SegmentationPseudo-labelingSelf-supervised SegmentationSelf-training
Deep learning models have demonstrated remarkable success in multi-organ segmentation but typically require large-scale datasets with all organs of interest annotated. However, medical image datasets are often low in sample size and only partially labeled, i.e., only a subset of organs are annotated. Therefore, it is crucial to investigate how to learn a unified model on the available partially labeled datasets to leverage their synergistic potential. In this paper, we systematically investigate the partial-label segmentation problem with theoretical and empirical analyses on the prior techniques. We revisit the problem from a perspective of partial label supervision signals and identify two signals derived from ground truth and one from pseudo labels. We propose a novel two-stage framework termed COSST, which effectively and efficiently integrates comprehensive supervision signals with self-training. Concretely, we first train an initial unified model using two ground truth-based signals and then iteratively incorporate the pseudo label signal to the initial model using self-training. To mitigate performance degradation caused by unreliable pseudo labels, we assess the reliability of pseudo labels via outlier detection in latent space and exclude the most unreliable pseudo labels from each self-training iteration. Extensive experiments are conducted on one public and three private partial-label segmentation tasks over 12 CT datasets. Experimental results show that our proposed COSST achieves significant improvement over the baseline method, i.e., individual networks trained on each partially labeled dataset. Compared to the state-of-the-art partial-label segmentation methods, COSST demonstrates consistent superior performance on various segmentation tasks and with different training data sizes.
32
ImageNet: A large-scale hierarchical image database
Jia Deng, Wei Dong, Richard Socher et al. · 2009 IEEE Conference on Computer Vision and Pattern Recognition · 2009 · 61.8K citations
Alexander M. Kirillov, Eric Mintun, Nikhila Ravi et al. · 2023 · 9.1K citations
The Medical Segmentation Decathlon
Michela Antonelli, Annika Reinke, Spyridon Bakas et al. · Nature Communications · 2022 · 1.2K citations · Full text
Semi-Supervised Semantic Segmentation with Cross Pseudo Supervision
Xiaokang Chen, Yuhui Yuan, Gang Zeng et al. · 2021 · 1.1K citations