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

The confounding effects of skin colour in photoacoustic imaging.

Nature communications·2026
Same author

SPEND-hSRS imaging of fumarate uncovers mitochondrial metabolic heterogeneity.

bioRxiv : the preprint server for biology·2026
Same author

Using Electrostatic Mapping to Understand PANI-MWCNTs' NH<sub>3</sub> Sensing.

Sensors (Basel, Switzerland)·2026
Same author

Early Radiation Therapy Response Assessment Using Multi-Scale Photoacoustic Imaging.

Advanced science (Weinheim, Baden-Wurttemberg, Germany)·2026
Same author

Optical coherence tomography and elastography for <i>ex vivo</i> visualization of early gastric cancer.

Journal of biomedical optics·2026
Same author

Harnessing Plant-Based Nanoparticles for Targeted Therapy: A Green Approach to Cancer and Bacterial Infections.

International journal of molecular sciences·2025

Related Experiment Video

Updated: Dec 27, 2025

Excitation-Scanning Hyperspectral Imaging Microscopy to Efficiently Discriminate Fluorescence Signals
07:34

Excitation-Scanning Hyperspectral Imaging Microscopy to Efficiently Discriminate Fluorescence Signals

Published on: August 22, 2019

8.3K

Deep learning applied to hyperspectral endoscopy for online spectral classification.

Alexandru Grigoroiu1,2, Jonghee Yoon1,2, Sarah E Bohndiek3,4

  • 1Department of Physics, University of Cambridge, JJ Thomson Avenue, Cambridge, CB3 0HE, United Kingdom.

Scientific Reports
|March 5, 2020
PubMed
Summary

A new convolutional neural network (CNN) enables real-time hyperspectral imaging (HSI) analysis during endoscopy. This AI tool accurately classifies colors in endoscopic images, aiding early cancer detection in the gastrointestinal tract.

More Related Videos

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

3.3K
Multimodal Imaging and Spectroscopy Fiber-bundle Microendoscopy Platform for Non-invasive, In Vivo Tissue Analysis
10:35

Multimodal Imaging and Spectroscopy Fiber-bundle Microendoscopy Platform for Non-invasive, In Vivo Tissue Analysis

Published on: October 17, 2016

8.2K

Related Experiment Videos

Last Updated: Dec 27, 2025

Excitation-Scanning Hyperspectral Imaging Microscopy to Efficiently Discriminate Fluorescence Signals
07:34

Excitation-Scanning Hyperspectral Imaging Microscopy to Efficiently Discriminate Fluorescence Signals

Published on: August 22, 2019

8.3K
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

3.3K
Multimodal Imaging and Spectroscopy Fiber-bundle Microendoscopy Platform for Non-invasive, In Vivo Tissue Analysis
10:35

Multimodal Imaging and Spectroscopy Fiber-bundle Microendoscopy Platform for Non-invasive, In Vivo Tissue Analysis

Published on: October 17, 2016

8.2K

Area of Science:

  • Medical imaging
  • Artificial intelligence
  • Gastroenterology

Background:

  • Hyperspectral imaging (HSI) offers biochemical information for early cancer detection in gastrointestinal endoscopy.
  • Real-time HSI deployment is advancing, but traditional analysis methods struggle with data volume for operator feedback.
  • Motion artifacts and data processing speed are key challenges in HSI endoscopy.

Purpose of the Study:

  • To develop and validate a convolutional neural network (CNN) for online classification of hyperspectral endoscopy data.
  • To enable real-time biochemical information extraction for improved cancer detection.
  • To overcome the limitations of traditional HSI analysis in processing high-volume data rapidly.

Main Methods:

  • A five-layered CNN was trained on 300 hyperspectral endoscopy images of a color chart.
  • The CNN was fine-tuned and tested on simulated endoscopic environments with warped color charts.
  • Performance was validated using ex vivo pig esophagus videos and human esophageal disease images.

Main Results:

  • The CNN achieved 94.3% accuracy in distinguishing 18 colors from a Macbeth ColorChecker chart at 8.8 fps.
  • The algorithm demonstrated robustness in simulated endoscopic conditions, maintaining over 90% accuracy.
  • Spatially distinct color classifications were observed in ex vivo and in vivo esophageal images.

Conclusions:

  • The proposed CNN shows potential for real-time, color-based classification in hyperspectral endoscopy.
  • This AI approach can aid in early cancer detection by providing rapid analysis of biochemical data.
  • Further validation is needed, but the results suggest a promising tool for endoscopic procedures.