RT Journal Article SR Electronic T1 Accelerating iterative deconvolution and multiview fusion by orders of magnitude JF bioRxiv FD Cold Spring Harbor Laboratory SP 647370 DO 10.1101/647370 A1 Min Guo A1 Yue Li A1 Yijun Su A1 Talley Lambert A1 Damian Dalle Nogare A1 Mark W. Moyle A1 Leighton H. Duncan A1 Richard Ikegami A1 Anthony Santella A1 Ivan Rey-Suarez A1 Daniel Green A1 Jiji Chen A1 Harshad Vishwasrao A1 Sundar Ganesan A1 Jennifer C. Waters A1 Christina M. Annunziata A1 Markus Hafner A1 William A. Mohler A1 Ajay B. Chitnis A1 Arpita Upadhyaya A1 Ted B. Usdin A1 Zhirong Bao A1 Daniel Colón-Ramos A1 Patrick La Riviere A1 Huafeng Liu A1 Yicong Wu A1 Hari Shroff YR 2019 UL http://biorxiv.org/content/early/2019/05/23/647370.abstract AB We describe theoretical and practical advances in algorithm and software design, resulting in ten to several thousand-fold faster deconvolution and multiview fusion than previous methods. First, we adapt methods from medical imaging, showing that an unmatched back projector accelerates Richardson-Lucy deconvolution by at least 10-fold, in most cases requiring only a single iteration. Second, we show that improvements in 3D image-based registration with GPU processing result in speedups of 10-100-fold over CPU processing. Third, we show that deep learning can provide further accelerations, particularly for deconvolution with a spatially varying point spread function. We illustrate the power of our methods from the subcellular to millimeter spatial scale, on diverse samples including single cells, nematode and zebrafish embryos, and cleared mouse tissue. Finally, we show that our methods facilitate the use of new microscopes that improve spatial resolution, including dual-view cleared tissue light-sheet microscopy and reflective lattice light-sheet microscopy.