Journal Article

Deep learning enables structured illumination microscopy with low light levels and enhanced speed

Liu, Bei, Jin, Luhong, Zhao, Fenqiang, Hahn, Stephen, Dong, Bowei, Song, Ruiyan, Elston, Timothy C., Hahn, Klaus M., Xu, Yingke

2020

Overview

Structured illumination microscopy (SIM) surpasses the optical diffraction limit and offers a two-fold enhancement in resolution over diffraction limited microscopy. However, it requires both intense illumination and multiple acquisitions to produce a single high-resolution image. Using deep learning to augment SIM, we obtain a five-fold reduction in the number of raw images required for super-resolution SIM, and generate images under extreme low light conditions (at least 100× fewer photons). We validate the performance of deep neural networks on different cellular structures and achieve multi-color, live-cell super-resolution imaging with greatly reduced photobleaching. Super-resolution microscopy typically requires high laser powers which can induce photobleaching and degrade image quality. Here the authors augment structured illumination microscopy (SIM) with deep learning to reduce the number of raw images required and boost its performance under low light conditions.

• Publisher: Nature Publishing Group UK,Nature Publishing Group,Nature Portfolio
• Subject: 14/63 / 631/114/1314 / 631/57/2267 / Animals / Artificial neural networks / Cellular communication / Cellular structure / Deep Learning / Diffraction / Fibroblasts - chemistry / Humanities and Social Sciences / Illumination / Image degradation / Image Processing, Computer-Assisted / Image quality / Image resolution / Light / Light levels / Machine learning / Mice / Microscopy / Microscopy - instrumentation / Microscopy - methods / multidisciplinary / Neural networks / Photobleaching / Photons / Science / Science (multidisciplinary)