Browse
Submit Data
Databases
API
Help

Dataset Information

0 Views

0 Connections

0 Citations

0 Reanalyses

0 Downloads

Omics score: 0

Deep learning for blind structured illumination microscopy.

ABSTRACT: Blind-structured illumination microscopy (blind-SIM) enhances the optical resolution without the requirement of nonlinear effects or pre-defined illumination patterns. It is thus advantageous in experimental conditions where toxicity or biological fluctuations are an issue. In this work, we introduce a custom convolutional neural network architecture for blind-SIM: BS-CNN. We show that BS-CNN outperforms other blind-SIM deconvolution algorithms providing a resolution improvement of 2.17 together with a very high Fidelity (artifacts reduction). Furthermore, BS-CNN proves to be robust in cross-database variability: it is trained on synthetically augmented open-source data and evaluated on experiments. This approach paves the way to the employment of CNN-based deconvolution in all scenarios in which a statistical model for the illumination is available while the specific realizations are unknown or noisy.

SUBMITTER: Xypakis E

PROVIDER: S-EPMC9124205 | biostudies-literature | 2022 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Json Xml

Publications

Deep learning for blind structured illumination microscopy.

Xypakis Emmanouil E Gosti Giorgio G Giordani Taira T Santagati Raffaele R Ruocco Giancarlo G Leonetti Marco M

Scientific reports 20220521 1

Blind-structured illumination microscopy (blind-SIM) enhances the optical resolution without the requirement of nonlinear effects or pre-defined illumination patterns. It is thus advantageous in experimental conditions where toxicity or biological fluctuations are an issue. In this work, we introduce a custom convolutional neural network architecture for blind-SIM: BS-CNN. We show that BS-CNN outperforms other blind-SIM deconvolution algorithms providing a resolution improvement of 2.17 together ...[more]

PMID: 35597874

Similar Datasets

Deep-MSIM: Fast Image Reconstruction with Deep Learning in Multifocal Structured Illumination Microscopy.

Project description:Fast and precise reconstruction algorithm is desired for for multifocal structured illumination microscopy (MSIM) to obtain the super-resolution image. This work proposes a deep convolutional neural network (CNN) to learn a direct mapping from raw MSIM images to super-resolution image, which takes advantage of the computational advances of deep learning to accelerate the reconstruction. The method is validated on diverse biological structures and in vivo imaging of zebrafish at a depth of 100 µm. The results show that high-quality, super-resolution images can be reconstructed in one-third of the runtime consumed by conventional MSIM method, without compromising spatial resolution. Last but not least, a fourfold reduction in the number of raw images required for reconstruction is achieved by using the same network architecture, yet with different training data.

| S-EPMC10520669 | biostudies-literature

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

Project description: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.

| S-EPMC7176720 | biostudies-literature

Structured line illumination Raman microscopy.

Project description:In the last couple of decades, the spatial resolution in optical microscopy has increased to unprecedented levels by exploiting the fluorescence properties of the probe. At about the same time, Raman imaging techniques have emerged as a way to image inherent chemical information in a sample without using fluorescent probes. However, in many applications, the achievable resolution is limited to about half the wavelength of excitation light. Here we report the use of structured illumination to increase the spatial resolution of label-free spontaneous Raman microscopy, generating highly detailed spatial contrast from the ensemble of molecular information in the sample. Using structured line illumination in slit-scanning Raman microscopy, we demonstrate a marked improvement in spatial resolution and show the applicability to a range of samples, including both biological and inorganic chemical component mapping. This technique is expected to contribute towards greater understanding of chemical component distributions in organic and inorganic materials.

| S-EPMC4686755 | biostudies-literature

Multifocal structured illumination optoacoustic microscopy.

Project description:Optoacoustic (OA) imaging has the capacity to effectively bridge the gap between macroscopic and microscopic realms in biological imaging. High-resolution OA microscopy has so far been performed via point-by-point scanning with a focused laser beam, thus greatly restricting the achievable imaging speed and/or field of view. Herein we introduce multifocal structured illumination OA microscopy (MSIOAM) that attains real-time 3D imaging speeds. For this purpose, the excitation laser beam is shaped to a grid of focused spots at the tissue surface by means of a beamsplitting diffraction grating and a condenser and is then scanned with an acousto-optic deflector operating at kHz rates. In both phantom and in vivo mouse experiments, a 10 mm wide volumetric field of view was imaged with 15 Hz frame rate at 28 μm spatial resolution. The proposed method is expected to greatly aid in biological investigations of dynamic functional, kinetic, and metabolic processes across multiple scales.

| S-EPMC7459102 | biostudies-literature

Machine learning assisted interferometric structured illumination microscopy for dynamic biological imaging.

Project description:Structured Illumination Microscopy, SIM, is one of the most powerful optical imaging methods available to visualize biological environments at subcellular resolution. Its limitations stem from a difficulty of imaging in multiple color channels at once, which reduces imaging speed. Furthermore, there is substantial experimental complexity in setting up SIM systems, preventing a widespread adoption. Here, we present Machine-learning Assisted, Interferometric Structured Illumination Microscopy, MAI-SIM, as an easy-to-implement method for live cell super-resolution imaging at high speed and in multiple colors. The instrument is based on an interferometer design in which illumination patterns are generated, rotated, and stepped in phase through movement of a single galvanometric mirror element. The design is robust, flexible, and works for all wavelengths. We complement the unique properties of the microscope with an open source machine-learning toolbox that permits real-time reconstructions to be performed, providing instant visualization of super-resolved images from live biological samples.

| S-EPMC9772218 | biostudies-literature

Projective oblique plane structured illumination microscopy.

Project description:Structured illumination microscopy (SIM) can double the spatial resolution of a fluorescence microscope and video rate live cell imaging in a two-dimensional format has been demonstrated. However, rapid implementations of 2D SIM typically only cover a narrow slice of the sample immediately at the coverslip, with most of the cellular volume out of reach. Here we implement oblique plane structured illumination microscopy (OPSIM) in a projection format to rapidly image an entire cell in a 2D SIM framework. As no mechanical scanning of the sample or objective is involved, this technique has the potential for rapid projection imaging with doubled resolution. We characterize the spatial resolution with fluorescent nanospheres, compare projection and 3D imaging using OPSIM and image mitochondria and ER dynamics across an entire cell at up to 2.7 Hz. To our knowledge, this represents the fastest whole cell SIM imaging to date.

| S-EPMC10441343 | biostudies-literature

Full-color structured illumination optical sectioning microscopy.

Project description:In merits of super-resolved resolution and fast speed of three-dimensional (3D) optical sectioning capability, structured illumination microscopy (SIM) has found variety of applications in biomedical imaging. So far, most SIM systems use monochrome CCD or CMOS cameras to acquire images and discard the natural color information of the specimens. Although multicolor integration scheme are employed, multiple excitation sources and detectors are required and the spectral information is limited to a few of wavelengths. Here, we report a new method for full-color SIM with a color digital camera. A data processing algorithm based on HSV (Hue, Saturation, and Value) color space is proposed, in which the recorded color raw images are processed in the Hue, Saturation, Value color channels, and then reconstructed to a 3D image with full color. We demonstrated some 3D optical sectioning results on samples such as mixed pollen grains, insects, micro-chips and the surface of coins. The presented technique is applicable to some circumstance where color information plays crucial roles, such as in materials science and surface morphology.

| S-EPMC4586488 | biostudies-literature

Cellular imaging of deep organ using two-photon Bessel light-sheet nonlinear structured illumination microscopy.

Project description:In vivo fluorescent cellular imaging of deep internal organs is highly challenging, because the excitation needs to penetrate through strong scattering tissue and the emission signal is degraded significantly by photon diffusion induced by tissue-scattering. We report that by combining two-photon Bessel light-sheet microscopy with nonlinear structured illumination microscopy (SIM), live samples up to 600 microns wide can be imaged by light-sheet microscopy with 500 microns penetration depth, and diffused background in deep tissue light-sheet imaging can be reduced to obtain clear images at cellular resolution in depth beyond 200 microns. We demonstrate in vivo two-color imaging of pronephric glomeruli and vasculature of zebrafish kidney, whose cellular structures located at the center of the fish body are revealed in high clarity by two-color two-photon Bessel light-sheet SIM.

| S-EPMC4026892 | biostudies-literature

Concepts for structured illumination microscopy with extended axial resolution through mirrored illumination.

Project description:Wide-field fluorescence microscopy, while much faster than confocal microscopy, suffers from a lack of optical sectioning and poor axial resolution. 3D structured illumination microscopy (SIM) has been demonstrated to provide optical sectioning and to double the resolution limit both laterally and axially, but even with this the axial resolution is still worse than the lateral resolution of unmodified wide-field microscopy. Interferometric schemes using two high numerical aperture objectives, such as 4Pi confocal and I5M microscopy, have improved the axial resolution beyond that of the lateral, but at the cost of a significantly more complex optical setup. Here, we theoretically and numerically investigate a simpler dual-objective scheme which we propose can be easily added to an existing 3D-SIM microscope, providing lateral and axial resolutions in excess of 125 nm with conventional fluorophores and without the need for interferometric detection.

| S-EPMC7173891 | biostudies-literature

Transmission structured illumination microscopy with tunable frequency illumination using tilt mirror assembly.

Project description:We present experimental demonstration of tilt-mirror assisted transmission structured illumination microscopy (tSIM) that offers a large field of view super resolution imaging. An assembly of custom-designed tilt-mirrors are employed as the illumination module where the sample is excited with the interference of two beams reflected from the opposite pair of mirror facets. Tunable frequency structured patterns are generated by changing the mirror-tilt angle and the hexagonal-symmetric arrangement is considered for the isotropic resolution in three orientations. Utilizing high numerical aperture (NA) objective in standard SIM provides super-resolution compromising with the field-of-view (FOV). Employing low NA (20X/0.4) objective lens detection, we experimentally demonstrate [Formula: see text] (0.56 mm[Formula: see text]0.35 mm) size single FOV image with [Formula: see text]1.7- and [Formula: see text]2.4-fold resolution improvement (exploiting various illumination by tuning tilt-mirrors) over the diffraction limit. The results are verified both for the fluorescent beads as well as biological samples. The tSIM geometry decouples the illumination and the collection light paths consequently enabling free change of the imaging objective lens without influencing the spatial frequency of the illumination pattern that are defined by the tilt-mirrors. The large and scalable FOV supported by tSIM will find usage for applications where scanning large areas are necessary as in pathology and applications where images must be correlated both in space and time.

| S-EPMC9879979 | biostudies-literature

OmicsDI is part of the ELIXIR infrastructure

OmicsDI is an Elixir interoperability service. Learn more ›

Tweets

OmicsDI Databases

PRIDE
PeptideAtlas
MassIVE
JPOST Repository
Physiome Model Repository

EGA
EVA
ENA
LINCS
PAXDB
Cell Collective

MetaboLights
Metabolomics Workbench
MetabolomeExpress
GNPS
BioModels
FAIRDOMHub

ArrayExpress
dbGaP
ExpressionAtlas
GEO
NODE

Information

Databases
Help
API
Contact us
Code on GitHub
Terms of use
Submit Data