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 4, 2026

Automated Joint Space Detection Improves Bone Segmentation Accuracy
06:45

Automated Joint Space Detection Improves Bone Segmentation Accuracy

Published on: November 28, 2025

Annotator reliability and probabilistic consensus for semantic segmentation in digital pathology.

Laura Gálvez Jiménez1, Christine Decaestecker2

  • 1Laboratory of Image Synthesis and Analysis, Université Libre de Bruxelles, Brussels, Belgium.

Artificial Intelligence in Medicine
|June 2, 2026
PubMed
Summary
This summary is machine-generated.

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

Decreased Expression of Aquaporins as a Feature of Tubular Damage in Lupus Nephritis.

Cells·2025
Same author

Inverse correlation between the amounts of lymphocytic infiltrate and stroma in breast carcinoma.

Heliyon·2024
Same author

Validation of a targeted next-generation sequencing panel for pancreatic ductal adenocarcinomas.

Experimental and molecular pathology·2024
Same author

Impact of imperfect annotations on CNN training and performance for instance segmentation and classification in digital pathology.

Computers in biology and medicine·2024
Same author

Correction: VEGFR1 and VEGFR2 Involvement in Extracellular Galectin-1- and Galectin-3-Induced Angiogenesis.

PloS one·2023
Same author

Usefulness of FAPα assessment in bronchoalveolar lavage as a marker of fibrogenesis: results of a preclinical study and first report in patients with idiopathic pulmonary fibrosis.

Respiratory research·2023

This study introduces a novel self-consistency method to quantify medical annotator certainty, improving machine learning model reliability in image segmentation. The approach enhances probabilistic consensus for better deep network training in digital pathology.

Area of Science:

  • Medical Image Analysis
  • Machine Learning
  • Digital Pathology

Background:

  • Expert annotations are crucial for training machine learning models in medical image segmentation.
  • Current consensus labels often fail to capture annotator consistency, impacting model reliability.
  • Variability in annotations, especially in cancer grading, poses a significant challenge.

Purpose of the Study:

  • To propose a novel, model-agnostic approach to quantify annotator certainty using self-consistency.
  • To improve the reliability of machine learning models by addressing annotation uncertainty.
  • To enhance the training of deep networks for semantic segmentation in medical imaging.

Main Methods:

  • Developed a self-consistency concept to characterize expert annotation behavior and quantify certainty.
Keywords:
Annotator reliabilityConsensusDigital pathologyMulti-expert annotationUncertainty

More Related Videos

Manual Segmentation of the Human Choroid Plexus Using Brain MRI
04:25

Manual Segmentation of the Human Choroid Plexus Using Brain MRI

Published on: December 15, 2023

Related Experiment Videos

Last Updated: Jun 4, 2026

Automated Joint Space Detection Improves Bone Segmentation Accuracy
06:45

Automated Joint Space Detection Improves Bone Segmentation Accuracy

Published on: November 28, 2025

Manual Segmentation of the Human Choroid Plexus Using Brain MRI
04:25

Manual Segmentation of the Human Choroid Plexus Using Brain MRI

Published on: December 15, 2023

  • Applied the method as a preprocessing step for segmentation backbones.
  • Validated the approach on the PANDA dataset with synthetic experts and real-world cancer datasets.

    • Main Results:

    • The self-consistency method provides valuable insights into intra-expert variability.
    • The approach improves the quality of probabilistic consensus compared to soft and smooth labeling.
    • Enhanced deep network training for semantic segmentation was observed.

    Conclusions:

    • The proposed self-consistency method effectively addresses annotation uncertainty in digital pathology.
    • This approach has the potential to improve the performance and reliability of AI models in medical image analysis.
    • The method offers a universal preprocessing step for various segmentation tasks.