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

SteuerLLM: local specialized large language model for German tax law analysis.

Scientific reports·2026
Same author

Inferring and evaluating network medicine-based disease modules with nextflow.

Bioinformatics (Oxford, England)·2026
Same author

Generating Alzheimer's narratives using large language models.

BMC medical informatics and decision making·2026
Same author

Pushing the boundaries of robotic computed tomography: automated twin-robot CT scan with maximum reachability.

Scientific reports·2026
Same author

Synthesizing vocal tract magnetic resonance imaging sequences with phoneme-aware diffusion models.

Journal of medical imaging (Bellingham, Wash.)·2026
Same author

Illuminating the black box of reservoir computing.

Scientific reports·2026

Related Experiment Video

Updated: Jun 2, 2025

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

42.8K

Attention-guided erasing for enhanced transfer learning in breast abnormality classification.

Adarsh Bhandary Panambur1,2, Sheethal Bhat3,4, Hui Yu4

  • 1Siemens Healthineers, Karl Heinz Kaske Str. 5, 91052, Erlangen, Bayern, Germany. adarsh.bhandary.panambur@fau.de.

International Journal of Computer Assisted Radiology and Surgery
|January 15, 2025
PubMed
Summary

Attention-Guided Erasing (AGE) improves breast cancer screening by enhancing deep learning models. This data augmentation technique boosts classification performance in mammography tasks, aiding early diagnosis.

Keywords:
Breast cancerData augmentationMammographySelf-supervised learningTransfer learning

More Related Videos

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.7K
Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
04:09

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

8.2K

Related Experiment Videos

Last Updated: Jun 2, 2025

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

42.8K
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.7K
Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
04:09

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

8.2K

Area of Science:

  • Medical Imaging Analysis
  • Artificial Intelligence in Healthcare
  • Machine Learning for Cancer Detection

Background:

  • Breast cancer is a leading global health concern, emphasizing the need for advanced screening and diagnostic tools.
  • Accurate classification of breast abnormalities in mammography is crucial for timely intervention and improved patient outcomes.
  • Transfer learning and data augmentation are key strategies for enhancing deep learning model performance in medical image analysis.

Purpose of the Study:

  • To evaluate the effectiveness and generalizability of the novel Attention-Guided Erasing (AGE) data augmentation technique.
  • To assess AGE's impact on various transfer learning classification tasks for breast abnormality detection in mammography.
  • To investigate the utility of self-supervised learning (DINO) for guiding data augmentation in medical imaging.

Main Methods:

  • Attention head visualizations from DINO self-supervised pretraining were used to identify regions of interest (ROIs).
  • The AGE technique stochastically erased non-essential background information in training images during transfer learning.
  • AGE was evaluated across five tasks: breast density (DM), malignancy (CEM), calcifications (SFM), masses (SFM), and ROI malignancy (CEM).

Main Results:

  • AGE significantly improved classification performance, with statistically significant gains in mean F1-scores across four of the five tasks.
  • Notable performance increases were observed for breast density classification (2%) and CEM malignancy classification (1.5%).
  • Minor improvements were seen in calcification classification (0.4% and 0.6%), with marginal gains in mass classification, suggesting task-specific optimization needs.

Conclusions:

  • Attention-Guided Erasing (AGE) is an effective dataset- and task-specific augmentation strategy for medical imaging.
  • AGE enhances downstream classification performance of deep learning models, particularly Vision Transformers (ViTs), in mammography analysis.
  • The integration of self-supervised learning with data augmentation shows significant promise for improving breast cancer screening accuracy.