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 Experiment Video

Updated: Dec 26, 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

43.5K

Few-shot learning with deformable convolution for multiscale lesion detection in mammography.

Ce Li1, Dong Zhang1, Zhiqiang Tian2

  • 1College of Electrical and Information Engineering, Lanzhou University of Technology, Lanzhou, 730050, China.

Medical Physics
|March 12, 2020
PubMed
Summary

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

2s-DAS: Two-Stream Diffusion with Multi-Modal Fusion for Temporal Action Segmentation.

Journal of imaging·2026
Same author

Integrating machine learning-based molecular design with experimental validation for the discovery of EGFR inhibitors in lung cancer.

Molecular diversity·2026
Same author

Ferroelectric-Polarization-Modulated 2D Floating-Gate Memory Enabling a 10<sup>6</sup> On/Off Ratio under ±1 V Gate-Voltage Sweep.

Nano letters·2026
Same author

Enriched Environment Suppresses Neuronal Ferroptosis Through SIRT1/AKT/GSK3β-Dependent Glycogen Metabolic Reprogramming After Cerebral Ischemia-Reperfusion.

Antioxidants (Basel, Switzerland)·2026
Same author

MS-PANet: Multi-Scale Spatial Pyramid Attention for Effective Drainage Pipeline Image Dehazing.

Journal of imaging·2026
Same author

Genomic Expansion and Adaptation in a Parasitoid Wasp <i>Eretmocerus hayati</i> (Hymenoptera: Aphelinidae).

Insects·2026

This study introduces FDMNet, a novel deep learning approach for breast lesion detection in mammography. By employing transfer learning (TL) and deformable convolution, FDMNet significantly improves detection accuracy, even with limited data, aiding physicians in diagnosis.

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Computer-Aided Diagnosis

Background:

  • Deep learning excels in medical image analysis but requires large datasets, which are scarce for mammography due to inconspicuous, multiscale lesions with blurred edges.
  • Conventional deep learning methods struggle with low detection accuracy in mammography owing to data limitations and lesion characteristics.

Purpose of the Study:

  • To enhance the accuracy of breast lesion detection in mammography for few-shot learning scenarios.
  • To develop a deep learning framework that assists physicians by improving mammary lesion detection capabilities.

Main Methods:

  • Proposed FDMNet (few-shot learning with deformable convolution for multiscale lesion detection in mammography).
  • Incorporated deformable convolution to improve lesion detection sensitivity.
Keywords:
attention factordeformable convolutionfew-shot learninglesion detectionmultiscale feature

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

3.3K

Related Experiment Videos

Last Updated: Dec 26, 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

43.5K
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

3.3K
  • Utilized a feature pyramid method to reinforce multiscale feature space sensitivity.
  • Integrated location information into the predictor to enhance lesion localization accuracy.
  • Employed transfer learning (TL) to transfer knowledge from a source domain to the target domain for improved few-shot learning performance.

    • Main Results:

    • FDMNet outperformed five conventional detection methods on public CBIS-DDSM and Mini-MIAS mammography datasets.
    • Achieved comprehensive scores of 0.911 (CBIS-DDSM) and 0.931 (Mini-MIAS).
    • Demonstrated high sensitivity (0.949 CBIS-DDSM, 0.966 Mini-MIAS) and precision (0.873 CBIS-DDSM, 0.882 Mini-MIAS).

    Conclusions:

    • The proposed method effectively addresses few-shot learning challenges in medical image analysis using TL for feature knowledge transformation.
    • The integration of deformable convolution and feature pyramid structure enhances network learning performance for lesion detection.
    • Comparative experiments confirm that FDMNet surpasses state-of-the-art methods in mammography lesion detection.