Deep learning for single-shot autofocus microscopy

Maintaining an in-focus image over long time scales is an essential and nontrivial task for a variety of microscopy applications. Here, we describe a fast, robust autofocusing method compatible with a wide range of existing microscopes. It requires only the addition of one or a few off-axis illumination sources (e.g., LEDs), and can predict the focus correction from a single image with this illumination. We designed a neural network architecture, the fully connected Fourier neural network (FCFNN), that exploits an understanding of the physics of the illumination to make accurate predictions with 2–3 orders of magnitude fewer learned parameters and less memory usage than existing state-of-the-art architectures, allowing it to be trained without any specialized hardware. We provide an open-source implementation of our method, to enable fast, inexpensive autofocus compatible with a variety of microscopes.

[1]  Shalin B. Mehta,et al.  Quantitative phase-gradient imaging at high resolution with asymmetric illumination-based differential phase contrast. , 2009, Optics letters.

[2]  M. Lohse,et al.  All-optical microscope autofocus based on an electrically tunable lens and a totally internally reflected IR laser. , 2018, Optics express.

[3]  L. Tian,et al.  Quantitative differential phase contrast imaging in an LED array microscope. , 2015, Optics express.

[4]  R. Horstmeyer,et al.  Wide-field, high-resolution Fourier ptychographic microscopy , 2013, Nature Photonics.

[5]  Laura Waller,et al.  Efficient illumination angle self-calibration in Fourier ptychography. , 2018, Applied optics.

[6]  Xin Zhang,et al.  Improvement in focusing accuracy of DNA sequencing microscope with multi-position laser differential confocal autofocus method. , 2018, Optics express.

[7]  Guoan Zheng,et al.  Quantitative phase imaging via Fourier ptychographic microscopy. , 2013, Optics letters.

[8]  Laura Waller,et al.  Multi-Contrast Imaging and Digital Refocusing on a Mobile Microscope with a Domed LED Array , 2015, PloS one.

[9]  Changhuei Yang,et al.  Microscopy refocusing and dark-field imaging by using a simple LED array. , 2011, Optics letters.

[10]  Andrew Zisserman,et al.  Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps , 2013, ICLR.

[11]  Laura Waller,et al.  Real-time brightfield, darkfield, and phase contrast imaging in a light-emitting diode array microscope , 2014, Journal of biomedical optics.

[12]  Paul C. Boutros,et al.  Germline contamination and leakage in whole genome somatic single nucleotide variant detection , 2017, bioRxiv.

[13]  Pascal Vincent,et al.  Stacked Denoising Autoencoders: Learning Useful Representations in a Deep Network with a Local Denoising Criterion , 2010, J. Mach. Learn. Res..

[14]  Stephan Hoyer,et al.  Assessing microscope image focus quality with deep learning , 2018, BMC Bioinformatics.

[15]  Edmund Y. Lam,et al.  Learning-based nonparametric autofocusing for digital holography , 2018 .

[16]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[17]  Klaus Hahn,et al.  Digital autofocus methods for automated microscopy. , 2006, Methods in enzymology.

[18]  Yibo Zhang,et al.  Extended depth-of-field in holographic image reconstruction using deep learning based auto-focusing and phase-recovery , 2018, Optica.

[19]  Nico Stuurman,et al.  Computer Control of Microscopes Using µManager , 2010, Current protocols in molecular biology.

[20]  Kaikai Guo,et al.  InstantScope: a low-cost whole slide imaging system with instant focal plane detection. , 2015, Biomedical optics express.

[21]  Kaikai Guo,et al.  Rapid focus map surveying for whole slide imaging with continuous sample motion , 2017 .

[22]  Zibang Zhang,et al.  Single-frame rapid autofocusing for brightfield and fluorescence whole slide imaging. , 2016, Biomedical optics express.

[23]  L. Tian,et al.  3D intensity and phase imaging from light field measurements in an LED array microscope , 2015 .

[24]  Henry Pinkard,et al.  Micro-Magellan: open-source, sample-adaptive, acquisition software for optical microscopy , 2016, Nature Methods.

[25]  M. Kreft,et al.  Focus‐Drift Correction in Time‐Lapse Confocal Imaging , 2005, Annals of the New York Academy of Sciences.

[26]  Laura Waller,et al.  Quasi-Dome: A Self-Calibrated High-NA LED Illuminator for Fourier Ptychography , 2017 .

[27]  Kaikai Guo,et al.  Rapid focus map surveying for whole slide imaging with continues sample motion , 2017, Optics letters.

[28]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[29]  Navid Farahani,et al.  A Practical Guide to Whole Slide Imaging: A White Paper From the Digital Pathology Association. , 2018, Archives of pathology & laboratory medicine.

[30]  Siavash Yazdanfar,et al.  Simple and robust image-based autofocusing for digital microscopy. , 2008, Optics express.