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

Weakly supervised regression enables interpretable tumor detection in whole-slide histopathology without negative cases.

Scientific reports·2025
Same author

Generating dermatopathology reports from gigapixel whole slide images with HistoGPT.

Nature communications·2025
Same author

Improving quality control of whole slide images by explicit artifact augmentation.

Scientific reports·2024
Same author

Electrical Compartmentalization in Neurons.

Cell reports·2019
Same author

Design and implementation of multi-signal and time-varying neural reconstructions.

Scientific data·2018
Same author

Cerebellar Nuclear Neurons Use Time and Rate Coding to Transmit Purkinje Neuron Pauses.

PLoS computational biology·2015

Related Experiment Video

Updated: Sep 4, 2025

Quantitation of Protein Expression and Co-localization Using Multiplexed Immuno-histochemical Staining and Multispectral Imaging
08:40

Quantitation of Protein Expression and Co-localization Using Multiplexed Immuno-histochemical Staining and Multispectral Imaging

Published on: April 8, 2016

12.9K

A Comparison Between Single- and Multi-Scale Approaches for Classification of Histopathology Images.

Marina D'Amato1, Przemysław Szostak2, Benjamin Torben-Nielsen1

  • 1Roche Information Solutions, F. Hoffmann-La Roche AG, Basel, Switzerland.

Frontiers in Public Health
|July 21, 2022
PubMed
Summary

Multi-scale analysis of whole slide images (WSIs) in digital pathology consistently improves classification performance. Combining multiple magnification levels enhances contextual and detailed information, leading to better diagnostic insights.

Keywords:
deep learningdigital pathologymulti-label classificationmulti-scale analysismultiple instance learningrepresentation learning

More Related Videos

Author Spotlight: Enhanced Multiplex Immunofluorescent Microscopy Protocol for Neuroscience Research
05:22

Author Spotlight: Enhanced Multiplex Immunofluorescent Microscopy Protocol for Neuroscience Research

Published on: June 21, 2024

516
Automated Quantification of Hematopoietic Cell – Stromal Cell Interactions in Histological Images of Undecalcified Bone
09:31

Automated Quantification of Hematopoietic Cell – Stromal Cell Interactions in Histological Images of Undecalcified Bone

Published on: April 8, 2015

11.7K

Related Experiment Videos

Last Updated: Sep 4, 2025

Quantitation of Protein Expression and Co-localization Using Multiplexed Immuno-histochemical Staining and Multispectral Imaging
08:40

Quantitation of Protein Expression and Co-localization Using Multiplexed Immuno-histochemical Staining and Multispectral Imaging

Published on: April 8, 2016

12.9K
Author Spotlight: Enhanced Multiplex Immunofluorescent Microscopy Protocol for Neuroscience Research
05:22

Author Spotlight: Enhanced Multiplex Immunofluorescent Microscopy Protocol for Neuroscience Research

Published on: June 21, 2024

516
Automated Quantification of Hematopoietic Cell – Stromal Cell Interactions in Histological Images of Undecalcified Bone
09:31

Automated Quantification of Hematopoietic Cell – Stromal Cell Interactions in Histological Images of Undecalcified Bone

Published on: April 8, 2015

11.7K

Area of Science:

  • Digital Pathology
  • Computational Histopathology
  • Medical Image Analysis

Background:

  • Whole slide images (WSIs) are high-resolution digitized histopathology slides.
  • WSIs are stored pyramidally, offering multiple magnification levels.
  • Current digital pathology algorithms often analyze WSIs at a single magnification, potentially missing crucial information.

Purpose of the Study:

  • To explore and compare multi-scale analysis approaches for WSIs in digital pathology.
  • To evaluate the effectiveness of multiple instance learning and a clustering-based representation learning (barcode) approach in a multi-scale setting.
  • To demonstrate the benefits of integrating information from different magnification levels for improved classification.

Main Methods:

  • Implemented and compared single-scale and multi-scale analyses.
  • Utilized a multiple instance learning framework.
  • Employed a representation learning algorithm (barcode approach) based on clustering.
  • Applied these methods to a multi-label histopathology classification task.

Main Results:

  • Multi-scale models consistently outperformed single-scale models.
  • Multiple instance learning achieved a 0.06 F1 score improvement in the multi-scale setting.
  • The barcode approach also demonstrated a 0.06 F1 score improvement with multi-scale analysis.
  • Consistent performance gains highlight the value of multi-scale WSI analysis.

Conclusions:

  • Multi-scale analysis of whole slide images (WSIs) is crucial for integrating global context and detailed spatial information.
  • Both multiple instance learning and the barcode approach benefit significantly from multi-scale WSI data.
  • Leveraging multiple magnification levels offers a promising avenue for enhancing diagnostic accuracy in digital pathology.