Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Efficient double helix detection with steerable filters.

Biomedical optics express·2026
Same author

Super-resolution atlas of SARS-CoV-2 infection reveals protease-dependent organelle maturation, dsRNA landscapes, and intracellular structural proteins.

Nature communications·2026
Same author

Laser phase plate improves structure determination of small proteins by cryo-EM.

Science (New York, N.Y.)·2026
Same author

Crossed laser phase plates for transmission electron microscopy.

Nature communications·2026
Same author

Glycan atlassing enables functional tracing of cell state.

Nature nanotechnology·2026
Same author

Organization of Myosin H in the Apical Complex of <i>Toxoplasma Gondii</i> Revealed by 3D Single-Molecule Super-Resolution Microscopy.

bioRxiv : the preprint server for biology·2026

Related Experiment Video

Updated: Jan 1, 2026

AMEBaS: Automatic Midline Extraction and Background Subtraction of Ratiometric Fluorescence Time-Lapses of Polarized Single Cells
06:03

AMEBaS: Automatic Midline Extraction and Background Subtraction of Ratiometric Fluorescence Time-Lapses of Polarized Single Cells

Published on: June 23, 2023

773

Accurate and rapid background estimation in single-molecule localization microscopy using the deep neural network

Leonhard Möckl1, Anish R Roy1, Petar N Petrov1

  • 1Department of Chemistry, Stanford University, Stanford, CA 94305.

Proceedings of the National Academy of Sciences of the United States of America
|December 25, 2019
PubMed
Summary

BGnet, a deep neural network, accurately estimates and removes structured background noise in microscopy images. This improves single-molecule localization precision and enhances super-resolution reconstruction quality.

Keywords:
background estimationdeep learninglocalization microscopysingle-molecule methodssuperresolution

More Related Videos

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

968
Localization and Relative Quantification of Carbon Nanotubes in Cells with Multispectral Imaging Flow Cytometry
14:09

Localization and Relative Quantification of Carbon Nanotubes in Cells with Multispectral Imaging Flow Cytometry

Published on: December 12, 2013

6.5K

Related Experiment Videos

Last Updated: Jan 1, 2026

AMEBaS: Automatic Midline Extraction and Background Subtraction of Ratiometric Fluorescence Time-Lapses of Polarized Single Cells
06:03

AMEBaS: Automatic Midline Extraction and Background Subtraction of Ratiometric Fluorescence Time-Lapses of Polarized Single Cells

Published on: June 23, 2023

773
Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

968
Localization and Relative Quantification of Carbon Nanotubes in Cells with Multispectral Imaging Flow Cytometry
14:09

Localization and Relative Quantification of Carbon Nanotubes in Cells with Multispectral Imaging Flow Cytometry

Published on: December 12, 2013

6.5K

Area of Science:

  • Optical Microscopy
  • Biophysics
  • Computational Imaging

Background:

  • Background fluorescence significantly reduces image quality in optical microscopy, particularly for single-molecule analysis.
  • Structured background noise is detrimental to 3D localization microscopy and single-molecule tracking.
  • Accurate background estimation is crucial for high-resolution imaging techniques.

Purpose of the Study:

  • To introduce BGnet, a deep neural network for rapid and accurate background estimation in microscopy images.
  • To demonstrate BGnet's effectiveness with various point-spread functions (PSFs) and complex background structures.
  • To show that BGnet improves localization precision and super-resolution reconstruction quality.

Main Methods:

  • Development of BGnet, a U-net-type deep neural network architecture.
  • Training BGnet on diverse simulated and experimental microscopy data with various PSFs.
  • Evaluation of BGnet's performance on different background structures and spatial frequencies.

Main Results:

  • BGnet accurately estimates background across a wide range of PSF shapes and background structures.
  • Background-corrected images show substantial improvement in localization precision for simulated and experimental data.
  • BGnet-based background estimation leads to higher quality super-resolution reconstructions of biological structures.

Conclusions:

  • BGnet provides a general and accurate method for structured background estimation in optical microscopy.
  • The developed deep learning approach significantly enhances the performance of single-molecule localization and super-resolution imaging.
  • BGnet is a valuable tool for improving image quality and data analysis in advanced microscopy applications.