Deep learning massively accelerates super-resolution localization microscopy

• Published in: Nature biotechnology Vol. 36; no. 5; pp. 460 - 468
• Main Authors: Ouyang, Wei, Aristov, Andrey, Lelek, Mickaël, Hao, Xian, Zimmer, Christophe
• Format: Journal Article
• Language: English
• Published: New York Nature Publishing Group US 01.06.2018; Nature Publishing Group
• Subjects: 13; 14; 14/1; 14/34; 14/63; 631/114/1305; 631/114/1564; 631/1647/245/2225; 631/1647/328; 631/1647/328/2238; Agriculture; Algorithms; Artificial neural networks; Bioinformatics; Biomedical Engineering/Biotechnology; Biomedicine; Biotechnology; Computer applications; Computer simulation; Deep Learning; Frames; Image acquisition; Image quality; Image reconstruction; Image resolution; Irradiation; Life Sciences; Localization; Methods; Microscopy; Microscopy - methods; Microscopy - statistics & numerical data; Microscopy, Fluorescence; Microscopy, Fluorescence - methods; Microscopy, Fluorescence - statistics & numerical data; Microtubules; Mitochondria; Molecular chains; Nanotechnology; Nanotechnology - methods; Nanotechnology - statistics & numerical data; Neural networks; Neural Networks (Computer); Nuclear pores; Palm; Radiation; Santé publique et épidémiologie; Spatial discrimination; Spatial resolution
• ISSN: 1087-0156; 1546-1696; 1546-1696
• DOI: 10.1038/nbt.4106

Accelerating PALM/STORM microscopy with deep learning allows super-resolution imaging of >1,000 cells in a few hours. The speed of super-resolution microscopy methods based on single-molecule localization, for example, PALM and STORM, is limited by the need to record many thousands of frames with a small number of observed molecules in each. Here, we present ANNA-PALM, a computational strategy that uses artificial neural networks to reconstruct super-resolution views from sparse, rapidly acquired localization images and/or widefield images. Simulations and experimental imaging of microtubules, nuclear pores, and mitochondria show that high-quality, super-resolution images can be reconstructed from up to two orders of magnitude fewer frames than usually needed, without compromising spatial resolution. Super-resolution reconstructions are even possible from widefield images alone, though adding localization data improves image quality. We demonstrate super-resolution imaging of >1,000 fields of view containing >1,000 cells in ∼3 h, yielding an image spanning spatial scales from ∼20 nm to ∼2 mm. The drastic reduction in acquisition time and sample irradiation afforded by ANNA-PALM enables faster and gentler high-throughput and live-cell super-resolution imaging.