美国国立卫生研究院Yicong Wu和浙江大学Huafeng Liu共同合作近期取得重要工作进展，他们研究发现将图像形成过程纳入深度学习可以提高网络性能。这一研究成果2022年10月31日在线发表于《自然—方法学》杂志上。

研究人员提出Richardson-Lucy网络（RLN），一种用于三维荧光显微镜去卷积快速和轻量级的深度学习方法。RLN将传统的Richardson-Lucy迭代与全卷积网络结构相结合，建立了与图像形成过程的联系，从而提高了网络性能。RLN只包含大约16,000个参数，与具有更多参数的纯数据驱动的网络相比，RLN的处理速度可以提高4到50倍。

通过视觉和定量分析，研究人员表明RLN提供了比其他网络更好的解卷积、更好的泛化能力和更少的伪影，特别是沿轴向维度。RLN在被严重失焦的荧光或噪声污染的体积上优于经典的Richardson-Lucy去卷积，并提供了比经典的多视图管道快4到6倍的大型已清除组织数据集的重建。研究人员展示了RLN在宽视场、光片、共聚焦和超分辨率显微镜成像的细胞、组织和胚胎上的表现。

附：英文原文

Title: Incorporating the image formation process into deep learning improves network performance

Author: Li, Yue, Su, Yijun, Guo, Min, Han, Xiaofei, Liu, Jiamin, Vishwasrao, Harshad D., Li, Xuesong, Christensen, Ryan, Sengupta, Titas, Moyle, Mark W., Rey-Suarez, Ivan, Chen, Jiji, Upadhyaya, Arpita, Usdin, Ted B., Coln-Ramos, Daniel Alfonso, Liu, Huafeng, Wu, Yicong, Shroff, Hari

Issue&Volume: 2022-10-31

Abstract: We present Richardson–Lucy network (RLN), a fast and lightweight deep learning method for three-dimensional fluorescence microscopy deconvolution. RLN combines the traditional Richardson–Lucy iteration with a fully convolutional network structure, establishing a connection to the image formation process and thereby improving network performance. Containing only roughly 16,000 parameters, RLN enables four- to 50-fold faster processing than purely data-driven networks with many more parameters. By visual and quantitative analysis, we show that RLN provides better deconvolution, better generalizability and fewer artifacts than other networks, especially along the axial dimension. RLN outperforms classic Richardson–Lucy deconvolution on volumes contaminated with severe out of focus fluorescence or noise and provides four- to sixfold faster reconstructions of large, cleared-tissue datasets than classic multi-view pipelines. We demonstrate RLN’s performance on cells, tissues and embryos imaged with widefield-, light-sheet-, confocal- and super-resolution microscopy.

DOI: 10.1038/s41592-022-01652-7

Source: https://www.nature.com/articles/s41592-022-01652-7