重新思考用于生物医学图像分割的 U-net 跳跃连接
Rethinking U-net Skip Connections for Biomedical Image Segmentation
AI总结：
本文研究 U-net 跳跃连接对域偏移的影响，发现浅层更易携带域特定信息，移除最上方跳跃连接可在域内和跨域分割中取得最佳性能。
摘要：
U-net 架构对基于深度学习的医学图像分割产生了重大影响。通过集成长程跳跃连接，它促进了高分辨率特征的保留。然而，分布外数据可能会显著阻碍神经网络的性能。先前的工作表明，训练后的网络各层对这种域偏移的易感性不同，例如，较浅层较深层受到的影响更大。在本工作中，我们研究了这一关于 U-net 风格分割网络各层对域偏移敏感性的观察所具有的意义。通过将浅层特征复制到相应的解码器块，这些跳跃连接存在重新引入特定域信息的风险。我们使用一个合成数据集来建模不同程度的数据分布偏移，并评估其对下游分割性能的影响。我们使用 Hellinger 距离量化了每个网络层固有的域易感性。这些实验证实了较早网络层具有更高的域易感性。当逐渐移除跳跃连接时，可以观察到较深层的域易感性降低。就下游分割性能而言，原始 U-net 优于没有任何跳跃连接的变体。然而，最佳性能是在移除最上方跳跃连接时取得的——不仅在存在域偏移的情况下如此，对于域内测试数据也是如此。我们在三个临床数据集上验证了结果——两个组织病理学数据集和一个磁共振数据集——当移除最上方跳跃连接时，域内性能提升最高达 10%，跨域性能提升最高达 13%。
英文摘要：
The U-net architecture has significantly impacted deep learning-based segmentation of medical images. Through the integration of long-range skip connections, it facilitated the preservation of high-resolution features. Out-of-distribution data can, however, substantially impede the performance of neural networks. Previous works showed that the trained network layers differ in their susceptibility to this domain shift, e.g., shallow layers are more affected than deeper layers. In this work, we investigate the implications of this observation of layer sensitivity to domain shifts of U-net-style segmentation networks. By copying features of shallow layers to corresponding decoder blocks, these bear the risk of re-introducing domain-specific information. We used a synthetic dataset to model different levels of data distribution shifts and evaluated the impact on downstream segmentation performance. We quantified the inherent domain susceptibility of each network layer, using the Hellinger distance. These experiments confirmed the higher domain susceptibility of earlier network layers. When gradually removing skip connections, a decrease in domain susceptibility of deeper layers could be observed. For downstream segmentation performance, the original U-net outperformed the variant without any skip connections. The best performance, however, was achieved when removing the uppermost skip connection - not only in the presence of domain shifts but also for in-domain test data. We validated our results on three clinical datasets - two histopathology datasets and one magnetic resonance dataset - with performance increases of up to 10% in-domain and 13% cross-domain when removing the uppermost skip connection.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
