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

Higher Levels of <i>BRCA1</i> Gene Methylation in Sporadic Breast Cancer Patients with a Lower Incidence of Recurrence.

Medical sciences (Basel, Switzerland)·2026
Same author

Refining HER2-Negative breast cancer: distinct survival profiles of HER2-Low and HER2-Ultralow tumors.

Pathology, research and practice·2025
Same author

Clinical, histopathological and immunohistochemical analysis of vulvar squamous cell carcinoma.

Revista brasileira de ginecologia e obstetricia : revista da Federacao Brasileira das Sociedades de Ginecologia e Obstetricia·2024
Same author

Molecular profile of gastric adenocarcinoma, relevant epidemiological factors - Systematic review and meta-analysis relating sex with Epstein-Barr virus and unstable microsatellites subtypes.

Asia-Pacific journal of clinical oncology·2023
Same author

Tumoral inflammatory infiltrate does not predict metastases in thin primary cutaneous melanomas.

Anais brasileiros de dermatologia·2023
Same author

Digital health in smart cities: Rethinking the remote health monitoring architecture on combining edge, fog, and cloud.

Health and technology·2023

Related Experiment Video

Updated: Apr 8, 2026

The Application of 1% Methylene Blue Dye As a Single Technique in Breast Cancer Sentinel Node Biopsy
07:51

The Application of 1% Methylene Blue Dye As a Single Technique in Breast Cancer Sentinel Node Biopsy

Published on: June 1, 2019

22.0K

MRNet: A Multi-Resolution Dual-Task Framework for Micrometastases Detection in Breast Cancer Sentinel Lymph Nodes.

Gabriela Kuhn1, João B Rodrigues Neto1, Felipe André Zeiser1,2,3

  • 1Universidade do Vale do Rio dos Sinos, Av. Unisinos, 950, 93022-000, São Leopoldo, RS, Brazil.

Journal of Imaging Informatics in Medicine
|April 6, 2026
PubMed
Summary

This study introduces MRNet, a deep learning framework improving breast cancer micrometastasis detection in lymph nodes. Its multi-resolution approach enhances accuracy and addresses limitations in current pathology diagnostics.

Keywords:
Breast cancerClassificationDeep learningHistopathologyMicrometastasesSegmentation

More Related Videos

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
07:15

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

7.7K
A Robust Discovery Platform for the Identification of Novel Mediators of Melanoma Metastasis
07:41

A Robust Discovery Platform for the Identification of Novel Mediators of Melanoma Metastasis

Published on: March 8, 2022

3.0K

Related Experiment Videos

Last Updated: Apr 8, 2026

The Application of 1% Methylene Blue Dye As a Single Technique in Breast Cancer Sentinel Node Biopsy
07:51

The Application of 1% Methylene Blue Dye As a Single Technique in Breast Cancer Sentinel Node Biopsy

Published on: June 1, 2019

22.0K
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
07:15

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

7.7K
A Robust Discovery Platform for the Identification of Novel Mediators of Melanoma Metastasis
07:41

A Robust Discovery Platform for the Identification of Novel Mediators of Melanoma Metastasis

Published on: March 8, 2022

3.0K

Area of Science:

  • Medical imaging
  • Computational pathology
  • Deep learning in oncology

Background:

  • Deep learning shows promise for detecting micrometastasis in breast cancer lymph nodes, aiding pathologists.
  • Current methods struggle with isolated tumor cells (low detection rates) and high false positives due to confusion with non-cancerous tissue.
  • Existing algorithms often use uniform resolution, failing to capture subtle features or handle annotation imprecision effectively.

Purpose of the Study:

  • To introduce MRNet, a novel multi-resolution dual-task framework to improve the detection of micrometastasis in breast cancer lymph nodes.
  • To address limitations of current deep learning methods, including poor performance on isolated tumor cells and high false-positive rates.
  • To develop a framework that optimizes resolution for different tasks (classification and segmentation) and handles annotation imprecision.

Main Methods:

  • MRNet employs a multi-resolution dual-task framework for analyzing gigapixel whole-slide images.
  • High-resolution patches (level 0) are used for classification to detect subtle micrometastatic features.
  • Moderate-resolution patches (level 3) are used for segmentation to mitigate annotation imprecision in histopathological datasets.

Main Results:

  • Achieved state-of-the-art classification performance with an area under the curve (AUC) of 0.998.
  • Reached a free-response operating characteristic (FROC) of 0.6124 in localization tasks.
  • Demonstrated that a multi-resolution strategy can systematically address the disconnect between patch-level and slide-level performance.

Conclusions:

  • MRNet offers a novel and effective multi-resolution approach for detecting breast cancer micrometastasis.
  • The framework significantly improves classification and localization performance compared to existing methods.
  • Resolution-aware design is crucial for overcoming limitations in deep learning for histopathological image analysis.