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 Experiment Video

Updated: Jun 25, 2026

Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

Comparison of deep learning approaches for extreme low-SNR image restoration.

Nasreen Elizabeth Buhn1, Sriya Reddy Adunur2, Joseph Hamilton3

  • 1Biological Sciences Department, California Polytechnic State University, 1 Grand Ave, San Luis Obispo, CA 93407-0401, USA.

Gigascience
|June 22, 2026
PubMed
Summary

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

Genetic or pharmacological disruption of the MSH3 Y245/K246 IDL binding pocket slows CAG repeat expansion.

NAR molecular medicine·2026
Same author

Correlative scanning electron and super-resolution structured illumination microscopy.

bioRxiv : the preprint server for biology·2026
Same author

Comparison of Deep Learning Approaches for Extreme Low-SNR Image Restoration.

bioRxiv : the preprint server for biology·2026
Same author

Journal Club Article: The STEROHCA trial - Optimizing post resuscitation haemodynamics by prehospital high dose corticosteroids.

Resuscitation plus·2025
Same author

Low-Cost Spinning Disk Confocal Microscopy with a 25-Megapixel Camera.

Sensors (Basel, Switzerland)·2025
Same author

A new family of bacterial actin-like proteins regulates cell morphology in a filamentous cyanobacterium.

mSphere·2025

A new fluorescence microscopy dataset and image stitching method address challenges in deep learning denoising. This benchmark aids in evaluating models for low-signal microscopy, with Transformer-based methods showing superior performance.

Area of Science:

  • Microscopy
  • Computational Imaging
  • Deep Learning

Background:

  • Live-cell fluorescence microscopy is vital for studying dynamic cellular processes.
  • Photobleaching and phototoxicity from microscopy can damage cells and disrupt processes.
  • Reducing light exposure yields low signal-to-noise ratio (SNR) images, hindering analysis.

Purpose of the Study:

  • To introduce a comprehensive dataset for evaluating deep learning denoising algorithms in fluorescence microscopy.
  • To address limitations in existing datasets and GPU memory constraints for processing large images.

Main Methods:

  • A novel fluorescence microscopy dataset with 17,568 paired high/low-SNR images across 15 diverse sub-datasets was created.
  • Five state-of-the-art deep learning denoising models (supervised, unsupervised, zero-shot) were evaluated.
Keywords:
deep learningdenoisingfluorescence microscopyimage restorationimage stitchingphototoxicity

More Related Videos

End-To-End Deep Neural Network for Salient Object Detection in Complex Environments
03:31

End-To-End Deep Neural Network for Salient Object Detection in Complex Environments

Published on: December 15, 2023

Related Experiment Videos

Last Updated: Jun 25, 2026

Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

End-To-End Deep Neural Network for Salient Object Detection in Complex Environments
03:31

End-To-End Deep Neural Network for Salient Object Detection in Complex Environments

Published on: December 15, 2023

  • An image stitching technique was developed to enable processing of large images in manageable crops.

    • Main Results:

    • The dataset offers a broad benchmark covering various specimens and imaging conditions.
    • The evaluated deep learning models demonstrated varying denoising capabilities.
    • The supervised Transformer-based model achieved the highest denoising performance.

    Conclusions:

    • The new dataset serves as a robust benchmark for deep learning denoising methods in microscopy.
    • The image stitching method effectively overcomes GPU memory limitations for large-scale image processing.
    • While Transformer-based models excel in denoising, they require significant training time.