Abstract
Structured Illumination Microscopy (SIM) is a widespread methodology to image live and fixed biological structures smaller than the diffraction limits of conventional optical microscopy. Using recent advances in image up-scaling through deep learning models, we demonstrate a method to reconstruct 3D SIM image stacks with twice the axial resolution attainable through conventional SIM reconstructions. We further demonstrate our method is robust to noise and evaluate it against two-point cases and axial gratings. Finally, we discuss potential adaptions of the method to further improve resolution.This article is part of the Theo Murphy meeting issue ‘Super-resolution structured illumination microscopy (part 1)’.
Highlights
Structured Illumination Microscopy (SIM) is a superresolution technique which enables a twofold increase in lateral resolution (X/Y axis, perpendicular to the line of sight of the microscope) when compared to conventional fluorescence microscopy [1,2]
Deep learning has been applied to SIM reconstruction with varying goals, such as accelerating the imaging process by reducing the required number of raw frames and inferring the spectral information from a smaller subset of SIM frames. These demonstrate an objective of high-throughput imaging, where a known microscope configuration is coupled with deep learning methods to minimize the imaging and processing time while improving the obtained resolution
The use of two simulated structure types in our training data demonstrates some capacity for the network to generalize to varying structures in observed specimens; this was reflected by the performance of the network on simulated Single Molecule Localization Microscopy (SMLM) data in figure 5
Summary
Structured Illumination Microscopy (SIM) is a superresolution technique which enables a twofold increase in lateral resolution (X/Y axis, perpendicular to the line of sight of the microscope) when compared to conventional fluorescence microscopy [1,2]. SIM functions via the illumination of structured light onto a specimen to obtain spectral information that would be out of the range visible to a wide-field microscope. A wide range of illumination patterns can be used, including 3D patterns that allow for improvements in axial resolution (Z axis, parallel to line of sight of the microscope) [3]. This has proven to be a useful investigative tool for the analysis of live biological processes as it avoids most sample damage traditionally associated with electron microscopy [4].
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
