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

Reply to the Letter to the Editor: Breast cancer risk prediction with artificial intelligence-statistical improvement versus clinical impact.

European radiology·2026
Same author

"Unanswered questions": Acceptability of a personalised breast cancer screening strategy in lower-risk women by healthcare professionals in the context of the MyPeBS study.

PloS one·2026
Same author

Anthropometric Characteristics of Triple-Negative Breast Cancer Patients by Menopausal Status: Evidence from the Population-Based Multicentric Study-MCC-Spain.

Healthcare (Basel, Switzerland)·2026
Same author

Biomarkers for Personalised Primary or Secondary Prevention in Cardiovascular Diseases: A Rapid Scoping Review.

International journal of molecular sciences·2025
Same author

Breast cancer risk assessment for screening: a hybrid artificial intelligence approach.

European radiology·2025
Same author

Object-Specific Multiview Classification Through View-Compatible Feature Fusion.

Sensors (Basel, Switzerland)·2025

Related Experiment Video

Updated: Aug 31, 2025

A Label-Free Segmentation Approach for Intravital Imaging of Mammary Tumor Microenvironment
10:39

A Label-Free Segmentation Approach for Intravital Imaging of Mammary Tumor Microenvironment

Published on: May 24, 2022

2.4K

Breast Dense Tissue Segmentation with Noisy Labels: A Hybrid Threshold-Based and Mask-Based Approach.

Andrés Larroza1, Francisco Javier Pérez-Benito1, Juan-Carlos Perez-Cortes1

  • 1Instituto Tecnológico de la Informática, Universitat Politècnica de València, Camino de Vera, 46022 València, Spain.

Diagnostics (Basel, Switzerland)
|August 26, 2022
PubMed
Summary
This summary is machine-generated.

A new deep learning method accurately estimates breast density from mammograms, outperforming expert radiologists. This automated approach improves breast cancer risk assessment by providing reliable breast density measurements.

Keywords:
breast density segmentationdeep learningmammographynoisy labels

More Related Videos

Area-based Image Analysis Algorithm for Quantification of Macrophage-fibroblast Cocultures
07:05

Area-based Image Analysis Algorithm for Quantification of Macrophage-fibroblast Cocultures

Published on: February 15, 2022

2.7K
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
13:44

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

43.0K

Related Experiment Videos

Last Updated: Aug 31, 2025

A Label-Free Segmentation Approach for Intravital Imaging of Mammary Tumor Microenvironment
10:39

A Label-Free Segmentation Approach for Intravital Imaging of Mammary Tumor Microenvironment

Published on: May 24, 2022

2.4K
Area-based Image Analysis Algorithm for Quantification of Macrophage-fibroblast Cocultures
07:05

Area-based Image Analysis Algorithm for Quantification of Macrophage-fibroblast Cocultures

Published on: February 15, 2022

2.7K
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
13:44

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

43.0K

Area of Science:

  • Radiology and Medical Imaging
  • Artificial Intelligence in Healthcare
  • Biomedical Engineering

Background:

  • Breast density on mammograms is a key breast cancer risk biomarker.
  • Current supervised learning methods for breast density assessment rely on expert labels, which are subject to intra- and inter-observer variability.
  • This variability introduces noise and limits the accuracy of breast density estimation.

Purpose of the Study:

  • To develop a fully automated deep learning method for accurate breast density estimation from digital mammograms.
  • To address the limitations of noisy expert labels by proposing a novel model that accounts for inter-reader variability.
  • To improve the reliability and consistency of breast density measurements for better breast cancer risk assessment.

Main Methods:

  • A novel confusion matrix (CM)-YNet deep learning model was developed for dense tissue segmentation.
  • The CM-YNet architecture incorporates networks to model individual radiologist labels, estimating ground truth and providing interactive threshold parameters.
  • A multi-center study utilized 1785 women's mammograms (2496 training, 844 testing) and an independent dataset (381 women) for validation.

Main Results:

  • The CM-YNet model achieved a superior average DICE score of 0.82±0.14 on test datasets, outperforming segmentation by individual radiologists (DICE 0.76±0.17).
  • The deep learning-based breast density estimator demonstrated higher performance compared to two experienced radiologists.
  • The model's ability to model radiologist variability suggests improved estimation of the true ground-truth segmentation.

Conclusions:

  • The proposed fully automated deep learning method offers a reliable and accurate approach to breast density estimation from mammograms.
  • Modeling individual radiologist variability enhances the accuracy of breast density segmentation, surpassing human expert performance.
  • The CM-YNet model's interactive threshold parameters offer potential for integration with existing clinical tools.