面向医学图像分割的带自适应深度监督的密集解码网络
Densely Decoded Networks with Adaptive Deep Supervision for Medical Image Segmentation
AI总结：
针对医学图像分割网络密集预测不足、特征鲁棒性弱的问题，提出带选择性辅助连接的密集解码网络与适配数据集属性的自适应深度监督训练策略，在4个多模态数据集上验证了有效性。
摘要：
基于深度神经网络的医学图像分割已取得显著成效，但这类网络的性能往往受限于密集预测能力不足、无法提取鲁棒特征的问题。为实现精细化的密集预测，我们提出密集解码网络（densely decoded networks，简称ddn），通过选择性引入“辅助”网络连接来优化性能。网络解码器每个上采样阶段的这类“辅助”连接，一方面通过融合编码器的高分辨率特征增强目标定位能力，另一方面通过促进多阶段上下文信息流动提升分割效果。此外，我们提出一种基于自适应深度监督（adaptive deep supervision，简称ads）的训练策略，该策略利用并适配输入数据集的特定属性，以实现鲁棒的特征提取。具体而言，ads通过将输入目标的平均尺寸与网络的逐层有效感受野（layer-wise effective receptive fields，简称lerf）相匹配，策略性地定位并部署辅助监督，由此构建出一类ddn网络。引入来自特定隐藏层的“伴随目标”，能帮助模型重点关注一些依赖输入的独特特征——若没有该机制，这些特征可能在训练过程中被网络“忽略”。我们在4个不同模态的多样化数据集上验证了所提出的新型网络与训练策略，证明了其有效性。
英文摘要：
Medical image segmentation using deep neural networks has been highly successful. However, the effectiveness of these networks is often limited by inadequate dense prediction and inability to extract robust features. To achieve refined dense prediction, we propose densely decoded networks (ddn), by selectively introducing 'crutch' network connections. Such 'crutch' connections in each upsampling stage of the network decoder (1) enhance target localization by incorporating high resolution features from the encoder, and (2) improve segmentation by facilitating multi-stage contextual information flow. Further, we present a training strategy based on adaptive deep supervision (ads), which exploits and adapts specific attributes of input dataset, for robust feature extraction. In particular, ads strategically locates and deploys auxiliary supervision, by matching the average input object size with the layer-wise effective receptive fields (lerf) of a network, resulting in a class of ddns. Such inclusion of 'companion objective' from a specific hidden layer, helps the model pay close attention to some distinct input-dependent features, which the network might otherwise 'ignore' during training. Our new networks and training strategy are validated on 4 diverse datasets of different modalities, demonstrating their effectiveness.
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。
