Deep Learning Microscopy: Enhancing Resolution, Field-of-View and Depth-of-Field of Optical Microscopy Images Using Neural Networks

By Yair Rivenson, Zoltan Gorocs, Harun Gunaydin, Yibo Zhang, Hongda Wang, Aydogan Ozcan. Conference on Lasers and Electro-Optics, 2017

403 citations recorded by the source corpus.

We demonstrate the ability of deep convolutional neural networks to significantly enhance the spatial resolution, field-of-view and depth-of-field of optical microscopy images, without any hardware modification to the imaging system.

Subfields: Atomic and Molecular Physics, and Optics, Biophysics

View Paper: https://www.semanticscholar.org/paper/cabe3d8a95563d787c7a9bebaa0b1f36fba923b1

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Atomic and Molecular Physics, and Optics

In Biophysics

Citations in the corpus

Probabilistic Representation and Inverse Design of Metamaterials Based on a Deep Generative Model with Semi‐Supervised Learning Strategy (2019, methodology): “…tool, DL has been applied to many other research fields such as material science [35, 36], chemistry [37-39], laser physics [40], particle physics [41], quantum mechanics [42, 43] and microscopy [44], showing great potential to circumvent the drawbacks of traditional methods in these areas.”
Content-aware image restoration: pushing the limits of fluorescence microscopy (2018, methodology): “In biology, DL methods have for instance been applied to the automatic extraction of connectomes from large electron microscopy data [28], for classification of image-based highcontent screens [29], fluorescence signal prediction from label-free images [30, 31], resolution enhancement in histopathol”
Deep learning for cellular image analysis (2019, background): “Other groups use deep learning to provide content-aware image denoising 146 , to improve image resolution 147 , and to mitigate axial undersampling 148 (to minimize phototoxicity) in real time.”
Applications, promises, and pitfalls of deep learning for fluorescence image reconstruction (2018, methodology): “To solve this problem, one could use lower-numeri-cal-aperture objectives that mechanically fit and apply deep learning to computationally increase the detection numerical aperture 45 .”
All-optical machine learning using diffractive deep neural networks (2018, methodology): “…function definition used in the training (see the Methods section) to teach the diffractive neural network to enhance the spatial resolution; in fact deep learning provides a powerful framework to improve image resolution by engineering the loss function used to train a neural network ( 8, 13 ).”
Inference in artificial intelligence with deep optics and photonics (2020, background): “Applications of deep learning in optical microscopy cover brightfield microscopy 80 , lensless microscopy 72,81,87 , fluorescence microscopy 82,83,85,86,88,89 , super-resolution microscopy 80,86,90–92 , confocal microscopy 85,86,93 , structured illumination microscopy 86 and many others 84,94–96 .”
Intelligent nanophotonics: merging photonics and artificial intelligence at the nanoscale (2018, background): “dinary capability in finding solutions from an enormous parameter space, researchers have started using deep learning in drug discovery [19], materials design [16,18,128], microscopy and spectroscopy [129, 130, 131, 132, 133, 134], and other physics-related domains [135, 136, 137]. Among all these a”