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

Super-resolution Fluorescence Microscopy01:37

Super-resolution Fluorescence Microscopy

7.0K
Super-resolution fluorescence microscopy (SRFM) provides a better resolution than conventional fluorescence microscopy by reducing the point spread function (PSF). PSF is the light intensity distribution from a point that causes it to appear blurred. Due to PSF, each fluorescing point appears bigger than its actual size, and it is the PSF interference of nearby fluorophores that causes the blurred image. Various approaches to achieving higher resolution through SRFM have recently been...
7.0K

You might also read

Related Articles

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

Sort by
Same author

Controlled encapsulation and droplet size prediction in two-step microfluidic double emulsions.

Lab on a chip·2026
Same author

Surfactant-laden droplet size prediction in a flow-focusing microchannel: a data-driven approach.

Lab on a chip·2022
See all related articles

Related Experiment Video

Updated: Jul 12, 2025

A Femtoliter Droplet Array for Massively Parallel Protein Synthesis from Single DNA Molecules
10:45

A Femtoliter Droplet Array for Massively Parallel Protein Synthesis from Single DNA Molecules

Published on: June 20, 2020

10.3K

Enhancing Microdroplet Image Analysis with Deep Learning.

Sofia H Gelado1, César Quilodrán-Casas2,3, Loïc Chagot4

  • 1Department of Computing, Imperial College London, London SW7 2AZ, UK.

Micromachines
|October 28, 2023
PubMed
Summary

Deep learning models like Segment Anything Model (SAM) enhance microfluidic imaging by improving droplet detection and measurement accuracy. These AI methods also enable super-resolution and denoising for clearer, more reliable microfluidic analyses.

Keywords:
computer visiondeep learningimage processingmicrodropletsmicrofluidics

More Related Videos

Author Spotlight: Unveiling the Polyfunctionality and Heterogeneity in Immune Responses
09:43

Author Spotlight: Unveiling the Polyfunctionality and Heterogeneity in Immune Responses

Published on: March 8, 2024

1.7K
DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning
04:17

DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning

Published on: May 10, 2024

790

Related Experiment Videos

Last Updated: Jul 12, 2025

A Femtoliter Droplet Array for Massively Parallel Protein Synthesis from Single DNA Molecules
10:45

A Femtoliter Droplet Array for Massively Parallel Protein Synthesis from Single DNA Molecules

Published on: June 20, 2020

10.3K
Author Spotlight: Unveiling the Polyfunctionality and Heterogeneity in Immune Responses
09:43

Author Spotlight: Unveiling the Polyfunctionality and Heterogeneity in Immune Responses

Published on: March 8, 2024

1.7K
DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning
04:17

DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning

Published on: May 10, 2024

790

Area of Science:

  • Microfluidics
  • Computer Vision
  • Artificial Intelligence

Background:

  • Microfluidics relies on precise imaging for process control and analysis.
  • Traditional image processing methods face limitations with low-quality or low-contrast microfluidic images.
  • Integrating deep learning offers potential for enhanced accuracy and automation in microfluidic image analysis.

Purpose of the Study:

  • To investigate the application of deep learning for accurate droplet detection and diameter measurement in microfluidics.
  • To evaluate deep learning-based image restoration techniques, including super-resolution and denoising, for microfluidic images.
  • To compare the performance of deep learning models against conventional methods in microfluidic image analysis.

Main Methods:

  • Utilized the Segment Anything Model (SAM) for droplet detection and diameter measurement, comparing it with the Circular Hough Transform.
  • Employed a deep learning super-resolution network (MSRN-BAM) trained on microfluidic droplet images for scales ×2, ×4, ×6, and ×8.
  • Applied a deep learning denoising model (DnCNN) to microfluidic images with additive Gaussian noise.

Main Results:

  • SAM demonstrated superior droplet detection and reduced diameter measurement errors compared to the Circular Hough Transform.
  • SAM showed increased robustness to image quality variations and low-contrast microfluidic images.
  • Super-resolved images achieved comparable detection and segmentation results to high-resolution images.
  • The DnCNN model effectively denoised microfluidic images with Gaussian noise (up to σ = 4).

Conclusions:

  • Deep learning methods, including SAM and MSRN-BAM, significantly improve the accuracy and reliability of microfluidic image analysis.
  • AI-powered image restoration techniques like super-resolution and denoising enhance the utility of low-quality microfluidic data.
  • Deep learning holds substantial potential for advancing various computer vision tasks within the field of microfluidics.