Su Jiawei, Luo Zhiming, Lian Sheng, Lin Dazhen, Li Shaozi
The Department of Artificial Intelligence, Xiamen University, Fujian, China.
The Department of Artificial Intelligence, Xiamen University, Fujian, China.
Med Image Anal. 2024 May;94:103111. doi: 10.1016/j.media.2024.103111. Epub 2024 Feb 21.
Semi-supervised learning has garnered significant interest as a method to alleviate the burden of data annotation. Recently, semi-supervised medical image segmentation has garnered significant interest that can alleviate the burden of densely annotated data. Substantial advancements have been achieved by integrating consistency-regularization and pseudo-labeling techniques. The quality of the pseudo-labels is crucial in this regard. Unreliable pseudo-labeling can result in the introduction of noise, leading the model to converge to suboptimal solutions. To address this issue, we propose learning from reliable pseudo-labels. In this paper, we tackle two critical questions in learning from reliable pseudo-labels: which pseudo-labels are reliable and how reliable are they? Specifically, we conduct a comparative analysis of two subnetworks to address both challenges. Initially, we compare the prediction confidence of the two subnetworks. A higher confidence score indicates a more reliable pseudo-label. Subsequently, we utilize intra-class similarity to assess the reliability of the pseudo-labels to address the second challenge. The greater the intra-class similarity of the predicted classes, the more reliable the pseudo-label. The subnetwork selectively incorporates knowledge imparted by the other subnetwork model, contingent on the reliability of the pseudo labels. By reducing the introduction of noise from unreliable pseudo-labels, we are able to improve the performance of segmentation. To demonstrate the superiority of our approach, we conducted an extensive set of experiments on three datasets: Left Atrium, Pancreas-CT and Brats-2019. The experimental results demonstrate that our approach achieves state-of-the-art performance. Code is available at: https://github.com/Jiawei0o0/mutual-learning-with-reliable-pseudo-labels.
半监督学习作为一种减轻数据标注负担的方法，已引起了广泛关注。最近，半监督医学图像分割也备受关注，它能够减轻密集标注数据的负担。通过整合一致性正则化和伪标签技术，已经取得了重大进展。在这方面，伪标签的质量至关重要。不可靠的伪标签会导致噪声的引入，使模型收敛到次优解。为了解决这个问题，我们提出从可靠的伪标签中学习。在本文中，我们解决了从可靠伪标签中学习的两个关键问题：哪些伪标签是可靠的，以及它们有多可靠？具体而言，我们对两个子网进行了比较分析，以应对这两个挑战。首先，我们比较两个子网的预测置信度。较高的置信度分数表明伪标签更可靠。随后，我们利用类内相似度来评估伪标签的可靠性，以应对第二个挑战。预测类别的类内相似度越高，伪标签就越可靠。子网根据伪标签的可靠性，有选择地纳入另一个子网模型传授的知识。通过减少不可靠伪标签带来的噪声引入，我们能够提高分割性能。为了证明我们方法的优越性，我们在三个数据集上进行了大量实验：左心房、胰腺CT和Brats - 2019。实验结果表明，我们的方法达到了当前最优性能。代码可在以下网址获取：https://github.com/Jiawei0o0/mutual-learning-with-reliable-pseudo-labels 。
