School of Computer Science and Engineering, Northeastern University, Shenyang, 110819, China.
School of Computer Science and Engineering, Northeastern University, Shenyang, 110819, China; Key Laboratory of Intelligent Computing in Medical Image, Ministry of Education, Northeastern University, Shenyang, 110819, China; Engineering Research Center of Security Technology of Complex Network System, Ministry of Education, China.
Comput Biol Med. 2023 May;157:106736. doi: 10.1016/j.compbiomed.2023.106736. Epub 2023 Mar 5.
Abundant labeled data drives the model training for better performance, but collecting sufficient labels is still challenging. To alleviate the pressure of label collection, semi-supervised learning merges unlabeled data into training process. However, the joining of unlabeled data (e.g., data from different hospitals with different acquisition parameters) will change the original distribution. Such a distribution shift leads to a perturbation in the training process, potentially leading to a confirmation bias. In this paper, we investigate distribution shift and develop methods to increase the robustness of our models, with the goal of improving performance in semi-supervised semantic segmentation of medical images. We study distribution shift and increase model robustness to it, for improving practical performance in semi-supervised segmentation over medical images.
To alleviate the issue of distribution shift, we introduce adversarial training into the co-training process. We simulate perturbations caused by the distribution shift via adversarial perturbations and introduce the adversarial perturbation to attack the supervised training to improve the robustness against the distribution shift. Benefiting from label guidance, supervised training does not collapse under adversarial attacks. For co-training, two sub-models are trained from two views (over two disjoint subsets of the dataset) to extract different kinds of knowledge independently. Co-training outperforms single-model by integrating both views of knowledge to avoid confirmation bias.
For practicality, we conduct extensive experiments on challenging medical datasets. Experimental results show desirable improvements to state-of-the-art counterparts (Yu and Wang, 2019; Peng et al., 2020; Perone et al., 2019). We achieve a DSC score of 87.37% with only 20% of labels on the ACDC dataset, almost same to using 100% of labels. On the SCGM dataset with more distribution shift, we achieve a DSC score of 78.65% with 6.5% of labels, surpassing 10.30% over Peng et al. (2020). Our evaluative results show superior robustness against distribution shifts in medical scenarios.
Empirical results show the effectiveness of our work for handling distribution shift in medical scenarios.
丰富的标记数据可推动模型训练，进而提升性能，但充分采集标签仍具挑战性。为缓解标签采集压力，半监督学习将未标记数据纳入训练过程。然而，未标记数据（例如，来自不同医院、具有不同采集参数的数据）的加入会改变原始分布。这种分布变化会导致训练过程中的干扰，可能导致确认偏倚。本文研究了分布转移，并开发了提高模型稳健性的方法，目标是提高医学图像半监督语义分割的性能。我们研究了分布转移并提高了模型对其的稳健性，以提高医学图像半监督分割的实际性能。
为缓解分布转移问题，我们在协同训练过程中引入对抗训练。我们通过对抗扰动模拟分布转移引起的扰动，并将对抗扰动引入到监督训练中，以提高对分布转移的稳健性。得益于标签指导，监督训练在对抗攻击下不会崩溃。对于协同训练，两个子模型从两个视图（在数据集的两个不相交子集上）进行训练，以独立提取不同类型的知识。协同训练通过整合两个视图的知识来避免确认偏倚，从而优于单个模型。
为了实用性，我们在具有挑战性的医学数据集上进行了广泛的实验。实验结果表明，与最先进的方法相比（Yu 和 Wang，2019；Peng 等人，2020；Perone 等人，2019），有了可喜的改进。在 ACDC 数据集上，仅使用 20%的标签，我们就达到了 87.37%的 DSC 评分，与使用 100%的标签几乎相同。在具有更多分布转移的 SCGM 数据集上，我们使用 6.5%的标签实现了 78.65%的 DSC 评分，超过了 Peng 等人（2020）的 10.30%。我们的评估结果表明，在医学场景中，我们的方法对分布转移具有更好的稳健性。
实验结果表明，我们的工作在处理医学场景中的分布转移方面是有效的。
