Incorporating the image formation process into deep learning improves network performance.

Journal: Nature methods
Published Date: Oct 31, 2022

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.

Authors

  • Yue Li
    School of Computer Science and Software Engineering, University of Science and Technology Liaoning, Anshan, China.
  • Yijun Su
    Laboratory of High Resolution Optical Imaging, National Institute of Biomedical Imaging and Bioengineering, National Institutes of Health, Bethesda, MD, USA.
  • Min Guo
    Key Laboratory of Biology and Sustainable Management of Plant Diseases and Pests of Anhui Higher Education Institutes, Hefei, People's Republic of China.
  • Xiaofei Han
    Laboratory of High Resolution Optical Imaging, National Institute of Biomedical Imaging and Bioengineering, National Institutes of Health, Bethesda, MD, USA.
  • Jiamin Liu
  • Harshad D Vishwasrao
    Advanced Imaging and Microscopy Resource, National Institutes of Health, Bethesda, MD, USA.
  • Xuesong Li
    Department of Chemistry, University of Wyoming, Laramie, WY, United States.
  • Ryan Christensen
    Laboratory of High Resolution Optical Imaging, National Institute of Biomedical Imaging and Bioengineering, National Institutes of Health, Bethesda, MD, USA.
  • Titas Sengupta
    Department of Neuroscience and Department of Cell Biology, Yale University School of Medicine, New Haven, CT, USA.
  • Mark W Moyle
    Department of Biology, Brigham Young University-Idaho, Rexburg, ID, USA.
  • Ivan Rey-Suarez
    Institute for Physical Science and Technology, University of Maryland, College Park, MD, USA.
  • Jiji Chen
    Advanced Imaging and Microscopy Resource, National Institutes of Health, Bethesda, MD, USA.
  • Arpita Upadhyaya
    Institute for Physical Science and Technology, University of Maryland, College Park, MD, USA.
  • Ted B Usdin
    Systems Neuroscience Imaging Resource, National Institute of Mental Health (NIMH), National Institutes of Health (NIH), Bethesda, MD, USA.
  • Daniel Alfonso Colón-Ramos
    Wu Tsai Institute, Department of Neuroscience and Department of Cell Biology, Yale University School of Medicine, New Haven, CT, USA.
  • Huafeng Liu
    State Key Laboratory of Modern Optical Instrumentation, Department of Optical Engineering, Zhejiang University, Hangzhou, China.
  • Yicong Wu
    Laboratory of High Resolution Optical Imaging, National Institute of Biomedical Imaging and Bioengineering, National Institutes of Health, Bethesda, MD, USA. yicong.wu@nih.gov.
  • Hari Shroff
    Laboratory of High Resolution Optical Imaging, National Institute of Biomedical Imaging and Bioengineering, National Institutes of Health, Bethesda, MD, USA.

Keywords

External Resources

Popular Topics
Recent Journals