Global voxel transformer networks for augmented microscopy
Global voxel transformer networks for augmented microscopy
复制标题
DOI：
10.1038/s42256-020-00283-x
复制
发表时间：
2020-08
影响因子：
23.8
通讯作者：
Zhengyang Wang;Yaochen Xie;Shuiwang Ji
中科院分区：
文献类型：
--
作者：
Zhengyang Wang;Yaochen Xie;Shuiwang Ji
深度学习的进步使增强显微镜技术取得了显著的成功，使我们能够在不使用昂贵的显微镜硬件和样品制备技术的情况下获得高质量的显微镜图像。目前用于增强显微镜的深度学习模型大多是基于U-Net的神经网络，因此具有某些限制性能的缺点。特别是，U-Nets仅由本地运营商组成，缺乏动态的非本地信息聚合。在这项工作中，我们引入了全局体素Transformer网络（GVTNets），这是一种用于增强显微镜的深度学习工具，它克服了当前基于U-Net的模型的固有局限性，并实现了性能的提高。GVTNet建立在全局体素Transformer运算符上，它能够聚合全局信息，而不是像卷积这样的局部运算符。我们将所提出的方法应用于现有的数据集上，在各种设置下进行三种不同的增强显微镜任务。
Advances in deep learning have led to remarkable success in augmented microscopy, enabling us to obtain high-quality microscope images without using expensive microscopy hardware and sample preparation techniques. Current deep learning models for augmented microscopy are mostly U-Net-based neural networks, thus sharing certain drawbacks that limit the performance. In particular, U-Nets are composed of local operators only and lack dynamic non-local information aggregation. In this work, we introduce global voxel transformer networks (GVTNets), a deep learning tool for augmented microscopy that overcomes intrinsic limitations of the current U-Net-based models and achieves improved performance. GVTNets are built on global voxel transformer operators, which are able to aggregate global information, as opposed to local operators like convolutions. We apply the proposed methods on existing datasets for three different augmented microscopy tasks under various settings.