Since its introduction, UNet has been leading a variety of medical image segmentation tasks. Although numerous follow-up studies have also been dedicated to improving the performance of standard UNet, few have conducted in-depth analyses of the underlying interest pattern of UNet in medical image segmentation. In this paper, we explore the patterns learned in a UNet and observe two important factors that potentially affect its performance: (i) irrelative feature learned caused by asymmetric supervision; (ii) feature redundancy in the feature map. To this end, we propose to balance the supervision between encoder and decoder and reduce the redundant information in the UNet. Specifically, we use the feature map that contains the most semantic information (i.e., the last layer of the decoder) to provide additional supervision to other blocks to provide additional supervision and reduce feature redundancy by leveraging feature distillation. The proposed method can be easily integrated into existing UNet architecture in a plug-and-play fashion with negligible computational cost. The experimental results suggest that the proposed method consistently improves the performance of standard UNets on four medical image segmentation datasets. The code is available at \url{https://github.com/ChongQingNoSubway/SelfReg-UNet}

翻译：自提出以来，UNet 一直在引领各类医学图像分割任务。尽管大量后续研究也致力于提升标准 UNet 的性能，但鲜有工作对其在医学图像分割中的内在兴趣模式进行深入分析。本文探讨了 UNet 学习到的模式，并观察到两个可能影响其性能的重要因素：(i) 由非对称监督导致学习到的无关特征；(ii) 特征图中的特征冗余。为此，我们提出平衡编码器与解码器之间的监督，并减少 UNet 中的冗余信息。具体而言，我们利用包含最丰富语义信息的特征图（即解码器的最后一层）为其他模块提供额外监督，并通过特征蒸馏来减少特征冗余。所提方法能够以即插即用的方式轻松集成到现有的 UNet 架构中，且计算开销可忽略不计。实验结果表明，该方法在四个医学图像分割数据集上均能持续提升标准 UNet 的性能。代码发布于 \url{https://github.com/ChongQingNoSubway/SelfReg-UNet}。