Field dependent deep learning enables high-throughput whole-cell 3D super-resolution imaging

Abstract

Single-molecule localization microscopy (SMLM) in a typical wide-field setup has been widely used for investigating sub-cellular structures with super resolution. However, field-dependent aberrations restrict the field of view (FOV) to only few tens of micrometers. Here, we present a deep learning method for precise localization of spatially variant point emitters (FD-DeepLoc) over a large FOV covering the full chip of a modern sCMOS camera. Using a graphic processing unit (GPU) based vectorial PSF fitter, we can fast and accurately model the spatially variant point spread function (PSF) of a high numerical aperture (NA) objective in the entire FOV. Combined with deformable mirror based optimal PSF engineering, we demonstrate high-accuracy 3D SMLM over a volume of ~180 × 180 × 5 μm3, allowing us to image mitochondria and nuclear pore complex in the entire cells in a single imaging cycle without hardware scanning - a 100-fold increase in throughput compared to the state-of-the-art.