半监督医学图像分割的跨金字塔一致性正则化
Cross-pyramid consistency regularization for semi-supervised medical image segmentation
AI总结：
针对半监督医学图像分割，本文提出结合跨金字塔一致性正则化（CPCR）的混合双分支金字塔网络（DBPNet），通过跨解码器金字塔预测的知识蒸馏利用未标记数据，在公共基准上性能优于五种先进自监督方法且与最新方法相当。
摘要：
半监督学习（SSL）能够在有限且精心标记的数据以及大量未标记数据支持学习的假设下训练强大模型。本文提出一种混合一致性学习方法，通过利用两个解码器之间的跨金字塔一致性正则化（CPCR），有效利用未标记数据进行半监督医学图像分割。首先，设计混合双分支金字塔网络（DBPNet），由一个编码器和两个略有不同的解码器组成，每个解码器在多个分辨率尺度上生成受扰动的辅助预测金字塔。其次，提出名为CPCR的网络学习策略，将现有一致性学习和不确定性最小化方法应用于解码器的主要输出预测，并结合新的正则化项。具体而言，该正则化项将软标签设置扩展到跨解码器的金字塔预测，以支持深度分层特征中的知识蒸馏。实验结果表明，带有CPCR的DBPNet在公共基准数据集上优于五种最先进的自监督学习方法，且性能与最新方法相当。
英文摘要：
Semi-supervised learning (SSL) enables training of powerful models with the assumption of limited, carefully labelled data and a large amount of unlabeled data to support the learning. In this paper, we propose a hybrid consistency learning approach to effectively exploit unlabeled data for semi-supervised medical image segmentation by leveraging Cross-Pyramid Consistency Regularization (CPCR) between two decoders. First, we design a hybrid Dual Branch Pyramid Network (DBPNet), consisting of an encoder and two decoders that differ slightly, each producing a pyramid of perturbed auxiliary predictions across multiple resolution scales. Second, we present a learning strategy for this network named CPCR that combines existing consistency learning and uncertainty minimization approaches on the main output predictions of decoders with our novel regularization term. More specifically, in this term, we extend the soft-labeling setting to pyramid predictions across decoders to support knowledge distillation in deep hierarchical features. Experimental results show that DBPNet with CPCR outperforms five state-of-the-art self-supervised learning methods and has comparable performance with recent ones on a public benchmark dataset.
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。
