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 Videos

Compound image segmentation of published biomedical figures.

Pengyuan Li1, Xiangying Jiang1, Chandra Kambhamettu1

  • 1Department of Computer and Information Sciences, University of Delaware, Newark, DE 19716, USA.

Bioinformatics (Oxford, England)
|October 18, 2017
PubMed
Summary

FigSplit effectively segments compound biomedical images into panels using Connected Component Analysis. This image segmentation system improves accuracy with a novel quality assessment and re-segmentation approach.

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

Anchor-based disentanglement framework for incremental multi-view clustering.

Neural networks : the official journal of the International Neural Network Society·2026
Same author

Comparative hyperparameter optimization of object detection models for precision monitoring of cucumber beetles and similar insects on yellow sticky cards.

Scientific reports·2026
Same author

AI in biocuration: challenges, opportunities, and a roadmap for sustainable integration.

Bioinformatics advances·2026
Same author

Plant-derived alginate polysaccharide hydrogels in sport and exercise nutrition: implications for carbohydrate metabolism, gastrointestinal integrity, exercise recovery, and athletic performance.

Frontiers in nutrition·2026
Same author

Effects of Nitrogen and Phosphorus Levels on Arbuscular Mycorrhizal Symbiosis and Associated Bacterial Communities in Culture.

Journal of fungi (Basel, Switzerland)·2025
Same author

Explainable artificial intelligence identifies and localizes left ventricular scar in hypertrophic cardiomyopathy using 12-Lead electrocardiogram.

Scientific reports·2025

Area of Science:

  • Biomedical image analysis
  • Bio-curation
  • Scientific data management

Background:

  • Biomedical publications increasingly use images as evidence for biological processes and experimental results.
  • Compound images, composed of multiple panels, are common but challenging to utilize due to the need for panel segmentation.
  • Automated segmentation of compound biomedical images is crucial for bio-curation and database development.

Purpose of the Study:

  • To develop an automated system for segmenting compound biomedical images into individual panels.
  • To enhance the accuracy and reliability of image segmentation through quality assessment and refinement.
  • To provide a robust solution for a critical step in utilizing visual data from biomedical literature.

Main Methods:

  • Development of the FigSplit system based on Connected Component Analysis.

Related Experiment Videos

  • Implementation of a quality assessment step to evaluate segmentation accuracy.
  • Introduction of two re-segmentation methods to correct inaccurate segmentations.

    • Main Results:

    • The FigSplit system demonstrates effective segmentation of compound biomedical images.
    • The integrated quality assessment and re-segmentation steps improve overall segmentation accuracy.
    • Experimental results confirm the superiority of FigSplit compared to existing methods.

    Conclusions:

    • FigSplit offers an effective solution for segmenting compound biomedical images.
    • The system's quality assessment and re-segmentation capabilities address limitations of previous approaches.
    • Automated segmentation of biomedical images facilitates better data curation and utilization.