Global voxel transformer networks for augmented microscopy

Advances in deep learning have led to remarkable success augmented microscopy, enabling us obtain high-quality microscope images without using expensive microscopy hardware and sample preparation techniques. Current models for are mostly U-Net-based neural networks, thus sharing certain drawbacks that limit the performance. In particular, U-Nets composed of local operators only lack dynamic non-local information aggregation. this work, we introduce global voxel transformer networks (GVTNets), a tool overcomes intrinsic limitations current achieves improved GVTNets built on operators, which able aggregate information, as opposed like convolutions. We apply proposed methods existing datasets three different tasks under various settings. Computational augmentation microscopic aims at reducing need chemically label or stain cells extract information. The popular U-Net model often employed these uses A new method augmenting is presented allows be used each step process.