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

Poly(ionic liquid)s: A Promising Matrix for Thermal Interface Materials.

ACS applied materials & interfaces·2024
Same author

Metal Ion Modulated Electron Transfer in Metalloviologen Compounds: Photochromism and Differentiable Amine Detection.

Inorganic chemistry·2024
Same author

Enhancing prostate cancer segmentation on multiparametric magnetic resonance imaging with background information and gland masks.

Medical physics·2024
Same author

Virtual reality technology reduces the pain and anxiety of children undergoing vein puncture: a meta-analysis.

BMC nursing·2024
Same author

Soft, Tough, Antifatigue Fracture Elastomer Composites with Low Thermal Resistance through Synergistic Crack Pinning and Interfacial Slippage.

Advanced materials (Deerfield Beach, Fla.)·2024
Same author

Soil quality enhancement by multi-treatment in the abandoned land of dry-hot river valley hydropower station construction area under karst desertification environment.

PloS one·2024

Related Experiment Video

Updated: Aug 9, 2025

Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging
15:48

Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging

Published on: December 15, 2014

22.6K

Weakly supervised breast lesion detection in DCE-MRI using self-transfer learning.

Rong Sun1, Xiaobing Zhang2, Yuanzhong Xie3

  • 1School of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai, China.

Medical Physics
|February 23, 2023
PubMed
Summary
This summary is machine-generated.

Weakly supervised deep learning effectively detects breast lesions in dynamic contrast-enhanced MRI (DCE-MRI). This approach reduces the need for extensive manual labeling, offering a promising tool for breast cancer diagnosis.

Keywords:
DCE-MRIbreast tumorlesion detectionself-transfer learningweakly supervised learning

More Related Videos

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

1.9K
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

6.9K

Related Experiment Videos

Last Updated: Aug 9, 2025

Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging
15:48

Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging

Published on: December 15, 2014

22.6K
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

1.9K
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

6.9K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Oncology

Background:

  • Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is crucial for breast lesion detection due to its high soft tissue resolution.
  • Supervised methods for breast lesion detection require significant time and expert staff for labeled training data.
  • There is a need for more efficient methods for breast lesion detection and diagnosis.

Purpose of the Study:

  • To evaluate the efficacy of weakly supervised deep learning models for detecting breast lesions in DCE-MRI.
  • To explore an automated approach for breast lesion detection, minimizing reliance on manual annotations.

Main Methods:

  • Trained a weakly supervised convolutional neural network (CNN) model on 1003 breast DCE-MRI studies using only image-level labels (normal/abnormal).
  • Simultaneously optimized the model for classification and detection sub-tasks.
  • Conducted ablation experiments to compare CNN backbones (VGG19 vs. ResNet50) and assess preprocessing effects.

Main Results:

  • The VGG19 backbone demonstrated superior performance compared to ResNet50 (p < 0.05).
  • Achieved high performance metrics: classification AP of 91.7% (abnormal) and 88.0% (normal), detection AP of 85.7%, and Correct Location (CorLoc) of 90.2%.
  • Area Under the ROC Curve (AUC) was 0.939, with a sensitivity of 84.0% at two false positives per image on FROC analysis.

Conclusions:

  • Weakly supervised CNNs utilizing self-transfer learning are effective auxiliary tools for breast lesion detection.
  • The proposed method shows promise in improving the efficiency and accuracy of breast cancer diagnosis using DCE-MRI.
  • This approach can potentially reduce the burden of manual data annotation in medical imaging AI.