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 20, 2026

High-Speed Ultraviolet Photoacoustic Microscopy for Histological Imaging with Virtual-Staining assisted by Deep Learning
09:31

High-Speed Ultraviolet Photoacoustic Microscopy for Histological Imaging with Virtual-Staining assisted by Deep Learning

Published on: April 28, 2022

UNICORN: a deep learning model for integrating multi-stain data in histopathology.

Valentin Koch1,2, Sabine Bauer3,4, Shweta Mahajan1,3

  • 1Computational Health Center, Helmholtz Munich - German Research Center for Environmental Health, Munich, Germany.

NPJ Digital Medicine
|June 18, 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

Illusion of competence: vision-language models provide confident but inaccurate explanations in cytological diagnostics.

Scientific reports·2026
Same author

Learning to See Peaks: Attention-Based Feature Extraction for Automated Chromatographic Peak Detection.

ACS omega·2026
Same author

AI-based hematological malignancy prediction from peripheral blood smears in a large diagnostic laboratory cohort.

Leukemia·2026
Same author

Weakly supervised deep learning for cutaneous squamous and basal cell carcinoma in whole-slide histopathology.

The journal of pathology. Clinical research·2026
Same author

EchoVisuALL: From Echocardiography to Gene Discovery.

bioRxiv : the preprint server for biology·2026
Same author

A disease model resource reveals core principles of tissue-specific cancer evolution.

Nature·2026

UNICORN, a novel deep learning model, effectively integrates multi-stain histopathology images for predicting atherosclerosis severity. This universal stain integration network outperforms existing methods by learning stain interactions, aiding disease progression analysis.

Area of Science:

  • Computational pathology
  • Artificial intelligence in medicine
  • Histopathology image analysis

Background:

  • Deep learning integration of multi-stain histopathology images presents challenges due to data heterogeneity and missing data.
  • Current methods struggle to effectively model stain-specific and cross-stain interactions when concatenating features.

Purpose of the Study:

  • To introduce UNICORN (UNiversal stain Integration network for CORonary classificatioN), a novel deep learning model for atherosclerosis severity prediction using multi-stain histopathology.
  • To address the limitations of existing approaches in handling data heterogeneity and learning stain interactions.

Main Methods:

  • UNICORN is a two-stage, end-to-end trainable model utilizing transformer self-attention blocks.
  • The first stage uses domain-specific expert models for feature extraction from individual stainings.

Related Experiment Videos

Last Updated: Jun 20, 2026

High-Speed Ultraviolet Photoacoustic Microscopy for Histological Imaging with Virtual-Staining assisted by Deep Learning
09:31

High-Speed Ultraviolet Photoacoustic Microscopy for Histological Imaging with Virtual-Staining assisted by Deep Learning

Published on: April 28, 2022

  • An aggregation expert model integrates these features by learning their interactions.

    • Main Results:

    • UNICORN achieved a classification accuracy of 0.68 on a multi-class, multi-stain whole slide images (WSIs) dataset of atherosclerotic lesions (MISSION dataset).
    • The model significantly outperformed state-of-the-art models in atherosclerosis severity prediction.
    • UNICORN demonstrated the ability to identify relevant tissue phenotypes across different stainings and implicitly model disease progression.

    Conclusions:

    • UNICORN offers an effective solution for integrating multi-stain histopathology images for disease prediction.
    • The model's explainability and accuracy highlight its potential for medical research and decision support in predicting atherosclerosis progression.
    • This approach shows promise for broader applications in analyzing complex histopathology data.