Artefact removal in ground truth deficient fluctuations-based nanoscopy images using deep learning

Abstract

Image denoising or artefact removal using deep learning is possible in the availability of supervised training dataset acquired in real experiments or synthesized using known noise models. Neither of the conditions can be fulfilled for nanoscopy (super-resolution optical microscopy) images that are generated from microscopy videos through statistical analysis techniques. Due to several physical constraints, a supervised dataset cannot be measured. Further, the non-linear spatio-temporal mixing of data and valuable statistics of fluctuations from fluorescent molecules that compete with noise statistics. Therefore, noise or artefact models in nanoscopy images cannot be explicitly learned. Here, we propose a robust and versatile simulation-supervised training approach of deep learning auto-encoder architectures for the highly challenging nanoscopy images of sub-cellular structures inside biological samples. We show the proof of concept for one nanoscopy method and investigate the scope of generalizability across structures, and nanoscopy algorithms not included during simulation-supervised training. We also investigate a variety of loss functions and learning models and discuss the limitation of existing performance metrics for nanoscopy images. We generate valuable insights for this highly challenging and unsolved problem in nanoscopy, and set the foundation for the application of deep learning problems in nanoscopy for life sciences.

Blind Super-Resolution Approach for Exploiting Illumination Variety in Optical-Lattice Illumination Microscopy

“…However, the postprocessing algorithms can be designed to compensate for such artifacts. 34 Figure 1e shows the radial line-scan of ground-truth, widefield, and final super-resolved image of the fluorescent sample. From the radial line scan profile, it is observed that separations more than 250 nm are fairly resolved, which signifies that we achieve around 3-fold lateral resolution enhancement over the diffraction limit (750 nm) using the sinusoidal lattice illumination.…”

“…The presence of such angle-specific artifacts indicates that this is primarily caused by the illumination rather than the reconstruction. However, the postprocessing algorithms can be designed to compensate for such artifacts Figure e shows the radial line-scan of ground-truth, widefield, and final super-resolved image of the fluorescent sample.…”

Supplementary document for Fluorescence fluctuation-based super-resolution microscopy using multimodal waveguided illumination - 5264968.pdf

“…The wide-field resolution of chip-based TIRFM and the potential of super-resolution through both single molecule localization and fluctuations based techniques have already been investigated [7][8][9]. At the same time, resolution and reconstruction quality of MUSICAL have also been studied previously [10,11]. The scope of utilizing the chip-based illumination for super-resolution using MUSICAL is new and the focus of the current article.…”