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

Modeling and Similitude01:12

Modeling and Similitude

Scaled modeling is a fundamental technique in engineering, enabling the study of large and complex systems by creating smaller, manageable replicas that recreate critical characteristics of the original. In hydrology and civil infrastructure, for example, scaled models of dams help analyze water flow, turbulence, and pressure. This method allows for accurate predictions of real-world behavior within a controlled environment, significantly reducing the cost and time involved in full-scale...
Causes of Similarity-Dissimilarity Effect01:26

Causes of Similarity-Dissimilarity Effect

The similarity-dissimilarity effect, a fundamental concept in social psychology, explains how interpersonal similarities and differences influence attraction and social interactions. This effect is supported by three key psychological perspectives: balance theory, social comparison theory, and consensual validation.Balance Theory and Cognitive ConsistencyBalance theory, developed by Fritz Heider, posits that individuals seek cognitive consistency in their relationships. When two people share...

You might also read

Related Articles

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

Sort by
Same author

Leveraging Advanced AI Frameworks for Dual PPAR α/γ Agonist Discovery in Alzheimer's Disease.

ACS chemical neuroscience·2026
Same author

Beyond Diagnosis: Evaluating Multimodal LLMs for Pathology Localization in Chest Radiographs.

Proceedings of machine learning research·2026
Same author

Leveraging Foundation Models for Histological Grading in Cutaneous Squamous Cell Carcinoma using PathFMTools.

Proceedings of machine learning research·2026
Same author

Development and validation of risk stratification models for hepatocellular cancer: A framework from the translational liver cancer consortium.

Hepatology (Baltimore, Md.)·2026
Same author

Integrating 730,947 exome sequences with clinical literature improves gene discovery.

medRxiv : the preprint server for health sciences·2026
Same author

Graph neural network modeling of spatial tumor-immune interactions identifies prognostic cellular niches in non‑small cell lung cancer.

NPJ precision oncology·2026

Related Experiment Video

Updated: Jul 10, 2026

Morphology-Based Distinction Between Healthy and Pathological Cells Utilizing Fourier Transforms and Self-Organizing Maps
08:59

Morphology-Based Distinction Between Healthy and Pathological Cells Utilizing Fourier Transforms and Self-Organizing Maps

Published on: October 28, 2018

Comparing Computational Pathology Foundation Models using Representational Similarity Analysis.

Vaibhav Mishra1, William Lotter2

  • 1Dana-Farber Cancer Institute.

Proceedings of Machine Learning Research
|July 9, 2026
PubMed
Summary

Computational pathology foundation models show distinct learned representations. Training methods impact model structure, with vision-language models having compact representations and high slide-dependence, which stain normalization can reduce.

Keywords:
computational pathologyfoundation modelsrepresentation analysisrobustness

More Related Videos

Three-Dimensional Shape Modeling and Analysis of Brain Structures
05:33

Three-Dimensional Shape Modeling and Analysis of Brain Structures

Published on: November 14, 2019

Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data
14:27

Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data

Published on: June 26, 2013

Related Experiment Videos

Last Updated: Jul 10, 2026

Morphology-Based Distinction Between Healthy and Pathological Cells Utilizing Fourier Transforms and Self-Organizing Maps
08:59

Morphology-Based Distinction Between Healthy and Pathological Cells Utilizing Fourier Transforms and Self-Organizing Maps

Published on: October 28, 2018

Three-Dimensional Shape Modeling and Analysis of Brain Structures
05:33

Three-Dimensional Shape Modeling and Analysis of Brain Structures

Published on: November 14, 2019

Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data
14:27

Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data

Published on: June 26, 2013

Area of Science:

  • Computational pathology
  • Computational neuroscience
  • Artificial intelligence in medicine

Background:

  • Foundation models are advancing computational pathology (CPath) for various tasks.
  • Understanding the learned representations of these models is crucial but underexplored.

Purpose of the Study:

  • To systematically analyze and compare the representational spaces of six CPath foundation models.
  • To investigate how training paradigms influence model representations and their variability.

Main Methods:

  • Representational similarity analysis applied to H&E image patches from TCGA.
  • Evaluation of six foundation models spanning vision-language contrastive learning and self-distillation.
  • Analysis of slide-dependence, disease-dependence, and intrinsic dimensionality.

Main Results:

  • UNI (v2) and Virchow (v2) exhibited the most distinct representational structures; Prov-GigaPath showed the highest average similarity.
  • Training paradigm (vision-only vs. vision-language) did not guarantee representational similarity.
  • All models displayed high slide-dependence, reduced by stain normalization, and low disease-dependence.
  • Vision-language models had more compact representations than vision-only models.

Conclusions:

  • Findings offer insights into model robustness, ensembling strategies, and the impact of training paradigms.
  • The developed framework can be extended to other medical imaging domains for foundation model development and deployment.