We present deep-learning-enabled super-resolution across different fluorescence microscopy modalities. This data-driven approach does not require numerical modeling of the imaging process or the estimation of a point-spread-function, and is based on training a generative adversarial network (GAN) to transform diffraction-limited input images into super-resolved ones. Using this framework, we improve the resolution of wide-field images acquired with low-numerical-aperture objectives, matching the resolution that is acquired using high-numerical-aperture objectives. We also demonstrate cross-modality super-resolution, transforming confocal microscopy images to match the resolution acquired with a stimulated emission depletion (STED) microscope. We further demonstrate that total internal reflection fluorescence (TIRF) microscopy images of subcellular structures within cells and tissues can be transformed to match the results obtained with a TIRF-based structured illumination microscope. The deep network rapidly outputs these super-resolved images, without any iterations or parameter search, and could serve to democratize super-resolution imaging.
This is a preview of subscription content
Subscribe to Nature+
Get immediate online access to the entire Nature family of 50+ journals
Subscribe to Journal
Get full journal access for 1 year
only $9.92 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Tax calculation will be finalised during checkout.
Get time limited or full article access on ReadCube.
All prices are NET prices.
We declare that all the data supporting the findings of this work are available within the manuscript and Supplementary Information files. Raw images can be requested from the corresponding author. Deep learning models reported in this work used standard libraries and scripts that are publicly available in TensorFlow. The instruction manual for our Fiji/ImageJ plugin and trained models (available online as Supplementary Software 1–7) is provided as a Supplementary Protocol.
Betzig, E. et al. Imaging intracellular fluorescent proteins at nanometer resolution. Science 313, 1642–1645 (2006).
Hess, S. T., Girirajan, T. P. K. & Mason, M. D. Ultra-high resolution imaging by fluorescence photoactivation localization microscopy. Biophys. J. 91, 4258–4272 (2006).
Rust, M. J., Bates, M. & Zhuang, X. Sub-diffraction-limit imaging by stochastic optical reconstruction microscopy (STORM). Nat. Methods 3, 793–795 (2006).
van de Linde, S. et al. Direct stochastic optical reconstruction microscopy with standard fluorescent probes. Nat. Protoc. 6, 991–1009 (2011).
Hell, S. W. & Wichmann, J. Breaking the diffraction resolution limit by stimulated emission: stimulated-emission-depletion fluorescence microscopy. Opt. Lett. 19, 780–782 (1994).
Gustafsson, M. G. L. Surpassing the lateral resolution limit by a factor of two using structured illumination microscopy. J. Microsc. 198, 82–87 (2000).
Cox, S. Super-resolution imaging in live cells. Dev. Biol. 401, 175–181 (2015).
Gustafsson, M. G. L. Nonlinear structured-illumination microscopy: wide-field fluorescence imaging with theoretically unlimited resolution. Proc. Natl. Acad. Sci. USA 102, 13081–13086 (2005).
Henriques, R. et al. QuickPALM: 3D real-time photoactivation nanoscopy image processing in ImageJ. Nat. Methods 7, 339–340 (2010).
Small, A. & Stahlheber, S. Fluorophore localization algorithms for super-resolution microscopy. Nat. Methods 11, 267–279 (2014).
Abraham, A. V., Ram, S., Chao, J., Ward, E. S. & Ober, R. J. Quantitative study of single molecule location estimation techniques. Opt. Express 17, 23352–23373 (2009).
Dempsey, G. T., Vaughan, J. C., Chen, K. H., Bates, M. & Zhuang, X. Evaluation of fluorophores for optimal performance in localization-based super-resolution imaging. Nat. Methods 8, 1027–1036 (2011).
Culley, S. et al. Quantitative mapping and minimization of super-resolution optical imaging artifacts. Nat. Methods 15, 263–266 (2018).
Sage, D. et al. Quantitative evaluation of software packages for single-molecule localization microscopy. Nat. Methods 12, 717–724 (2015).
Almada, P., Culley, S. & Henriques, R. PALM and STORM: into large fields and high-throughput microscopy with sCMOS detectors. Methods 88, 109–121 (2015).
Goodfellow, I. J. et al. Generative adversarial networks. arXiv Preprint at https://arxiv.org/abs/1406.2661 (2014).
Wilson, T. & Masters, B. R. Confocal microscopy. Appl. Opt. 33, 565–566 (1994).
Li, D. et al. Extended-resolution structured illumination imaging of endocytic and cytoskeletal dynamics. Science 349, aab3500 (2015).
Richardson, W. H. Bayesian-based iterative method of image restoration. J. Opt. Soc. Am. 62, 55 (1972).
Lucy, L. B. An iterative technique for the rectification of observed distributions. Astron. J. 79, 745 (1974).
Landweber, L. An iteration formula for Fredholm integral equations of the first kind. Am. J. Math. 73, 615–624 (1951).
Farahani, J. N., Schibler, M. J. & Bentolila, L. A. Stimulated emission depletion (STED)microscopy: from theory to practice. Microsc. Sci. Technol. Appl. Educ. 2, 1539–1547 (2010).
Hamel, P., Davies, M. E. P., Yoshii, K. & Goto, M. Transfer learning in MIR: sharing learned latent representations for music audio classification and similarity. Google AI https://ai.google/research/pubs/pub41530 (2013).
Wäldchen, S., Lehmann, J., Klein, T., van de Linde, S. & Sauer, M. Light-induced cell damage in live-cell super-resolution microscopy. Sci. Rep. 5, 15348 (2015).
Hein, B., Willig, K. I. & Hell, S. W. Stimulated emission depletion (STED) nanoscopy of a fluorescent protein-labeled organelle inside a living cell. Proc. Natl. Acad. Sci. USA 105, 14271–14276 (2008).
Hein, B. et al. Stimulated emission depletion nanoscopy of living cells using SNAP-tag fusion proteins. Biophys. J. 98, 158–163 (2010).
Dyba, M. & Hell, S. W. Photostability of a fluorescent marker under pulsed excited-state depletion through stimulated emission. Appl. Opt. 42, 5123–5129 (2003).
Kner, P., Chhun, B. B., Griffis, E. R., Winoto, L. & Gustafsson, M. G. L. Super-resolution video microscopy of live cells by structured illumination. Nat. Methods 6, 339–342 (2009).
Leyton-Puig, D. et al. Flat clathrin lattices are dynamic actin-controlled hubs for clathrin-mediated endocytosis and signalling of specific receptors. Nat. Commun. 8, 16068 (2017).
Fiolka, R., Shao, L., Rego, E. H., Davidson, M. W. & Gustafsson, M. G. L. Time-lapse two-color 3D imaging of live cells with doubled resolution using structured illumination. Proc. Natl. Acad. Sci. USA 109, 5311–5315 (2012).
Ferguson, J. P. et al. Deciphering dynamics of clathrin-mediated endocytosis in a living organism. J. Cell. Biol. 214, 347–358 (2016).
Forster, B. et al. M. Complex wavelets for extended depth-of-field: a new method for the fusion of multichannel microscopy images. Microsc. Res. Tech. 65, 33–42 (2004).
Liu, R. & Jia, J. Reducing boundary artifacts in image deconvolution. in 2008 15th IEEE International Conference on Image Processing 505–508 (IEEE, New York, 2008).
Cox, I. J. & Sheppard, C. J. R. Information capacity and resolution in an optical system. J. Opt. Soc. Am. A. 3, 1152–1158 (1986).
Katznelson, Y. An Introduction to Harmonic Analysis (Dover Publications, New York, 1976).
Bentolila, L. A. et al. Imaging of angiotropism/vascular co-option in a murine model of brain melanoma: implications for melanoma progression along extravascular pathways. Sci. Rep. 6, 23834 (2016).
Aguet, F. et al. Membrane dynamics of dividing cells imaged by lattice light-sheet microscopy. Mol. Biol. Cell 27, 3418–3435 (2016).
Willy, N. M. et al. Membrane mechanics govern spatiotemporal heterogeneity of endocytic clathrin coat dynamics. Mol. Biol. Cell 28, 3480–3488 (2017).
Schindelin, J. et al. Fiji: an open-source platform for biological-image analysis. Nat. Methods 9, 676–682 (2012).
Preibisch, S., Saalfeld, S. & Tomancak, P. Globally optimal stitching of tiled 3D microscopic image acquisitions. Bioinformatics 25, 1463–1465 (2009).
Sage, D., Prodanov, D., Tinevez, J.-Y. & Schindelin, J. MIJ: making interoperability between ImageJ and Matlab possible. Poster presented at the ImageJ User & Developer Conference, Mondorf-les-Bains, Luxembourg, 24–26 October, 2012.
Rivenson, Y. et al. Deep learning enhanced mobile-phone microscopy. ACS Photonics 5, 2354–2364 (2018).
Rivenson, Y. et al. Deep learning-based virtual histology staining using auto-fluorescence of label-free tissue. arXiv Preprint at https://arxiv.org/abs/1803.11293 (2018).
Wang, Z., Bovik, A. C., Sheikh, H. R. & Simoncelli, E. P. Image quality assessment: from error visibility to structural similarity. IEEE Trans. Image Process. 13, 600–612 (2004).
Ronneberger, O., Fischer, P. & Brox, T. U-Net: convolutional networks for biomedical image segmentation. arXiv Preprint at https://arxiv.org/abs/1505.04597 (2015).
Wu, Y. et al. Extended depth-of-field in holographic imaging using deep-learning-based autofocusing and phase recovery. Optica 5, 704–710 (2018).
Kingma, D. P. & Ba, J. Adam: a method for stochastic optimization. arXiv Preprint at https://arxiv.org/abs/1412.6980 (2014).
Abadi, M. et al. TensorFlow: a system for large-scale machine learning. arXiv Preprint at https://arxiv.org/abs/1605.08695 (2016).
Aguet, F., Van De Ville, D. & Unser, M. Model-based 2.5-d deconvolution for extended depth of field in brightfield microscopy. IEEE Trans. Image Process. 17, 1144–1153 (2008).
Born, M., Wolf, E. & Bhatia, A. B. Principles of Optics: Electromagnetic Theory of Propagation, Interference and Diffraction of Light (Cambridge University Press, 1999).
Kirshner, H., Aguet, F., Sage, D. & Unser, M. 3-D PSF fitting for fluorescence microscopy: implementation and localization application. J. Microsc. 249, 13–25 (2013).
Sage, D. et al. DeconvolutionLab2: an open-source software for deconvolution microscopy. Methods 115, 28–41 (2017).
The Ozcan Research Group at UCLA acknowledges the support of NSF Engineering Research Center (ERC, PATHS-UP), the Army Research Office (ARO; W911NF-13-1-0419 and W911NF-13-1-0197), the ARO Life Sciences Division, the National Science Foundation (NSF) CBET Division Biophotonics Program, the NSF Emerging Frontiers in Research and Innovation (EFRI) Award, the NSF INSPIRE Award, NSF Partnerships for Innovation: Building Innovation Capacity (PFI:BIC) Program, the National Institutes of Health (NIH, R21EB023115), the Howard Hughes Medical Institute (HHMI), Vodafone Americas Foundation, the Mary Kay Foundation, and Steven & Alexandra Cohen Foundation. Yair Rivenson is partially supported by the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No H2020-MSCA-IF-2014-659595 (MCMQCT). Confocal and STED laser scanning microscopy was performed at the California NanoSystems Institute (CNSI) Advanced Light Microscopy/Spectroscopy Shared Resource Facility at UCLA. We also thank the Advanced Imaging Center (AIC) at Janelia Research Campus for access to their TIRF-SIM microscope. The AIC is jointly supported by the Howard Hughes Medical Institute and the Gordon and Betty Moore Foundation. Finally, we thank H. Chang (Purdue University, West Lafayette, IN, USA) for sharing the CLC-mEmerald fly strain.
A.O., Y.R., and H.W. have a pending patent application on the contents of the presented results.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Notes 1–10, Supplementary Figures 1–14 and Supplementary Table 1
SISR-Fluorescent Fiji/ImageJ plugin: User Manual
Deep-learning enabled cross-modality image transformation from TIRF to TIRF-SIM. A video of deep-learning-enabled cross-modality image transformation from TIRF to TIRF-SIM, corresponding to a gene-edited SUM159 cell expressing AP2-eGFP, revealing the temporal dynamics of endocytic protein structures within the cell. The highlighted frames in this video correspond to subpanels of Fig. 6 (main text). Experiments were repeated with >1,000 images/frames, achieving similar results.
SISR-Fluorescent Fiji/ImageJ plugin
Pre-trained model: wide-field DAPI
Pre-trained model: wide-field FITC
Pre-trained model: wide-field TxRed
Pre-trained model: confocal STED (nanobeads)
Pre-trained model: confocal STED (nuclei)
Pre-trained model: TIRF-SIM
About this article
Cite this article
Wang, H., Rivenson, Y., Jin, Y. et al. Deep learning enables cross-modality super-resolution in fluorescence microscopy. Nat Methods 16, 103–110 (2019). https://doi.org/10.1038/s41592-018-0239-0
Eye and Vision (2022)
Light: Science & Applications (2022)
Generative adversarial network enables rapid and robust fluorescence lifetime image analysis in live cells
Communications Biology (2022)
Scientific Reports (2022)