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

Serum Pharmacochemistry-Guided DARTS-MS Profiling Reveals Potential Mechanisms of <i>Caragana jubata</i> Against Hypoxic Pulmonary Hypertension.

International journal of molecular sciences·2026
Same author

Co-occurring polycyclic aromatic hydrocarbons and heavy metals drive bacterial community shifts regulated by soil water content in China's Beiluo River riparian soils.

Ecotoxicology (London, England)·2026
Same author

Adaptive texture and low-light feature learning in enhanced MobileNetV4 for industrial packaging quality inspection.

Scientific reports·2026
Same author

Distinct associations between polycyclic aromatic hydrocarbons with different molecular weights and antibiotic resistance gene distribution in river sediments of the Loess Plateau, China.

Environmental research·2026
Same author

A network analysis of symptom clusters and core symptoms in colorectal cancer patients undergoing postoperative chemotherapy.

Frontiers in oncology·2026
Same author

Hyperbolic Shear Plasmon Polaritons in a Twisted Anisotropic 2D Material.

Physical review letters·2026

Related Experiment Video

Updated: Oct 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

43.1K

Classification of microcalcification clusters in digital breast tomosynthesis using ensemble convolutional neural

Bingbing Xiao1, Haotian Sun2,3, You Meng4,5

  • 1Institute of Biomedical Engineering, School of Communication and Information Engineering, Shanghai University, Shanghai, China.

Biomedical Engineering Online
|July 29, 2021
PubMed
Summary

This study introduces an ensemble deep learning model for classifying microcalcification clusters in digital breast tomosynthesis (DBT) images. The novel approach significantly improves accuracy and reduces false positives in diagnosing breast cancer.

Keywords:
ClassificationConvolution neural networkDigital breast tomosynthesisEnsemble learningMicrocalcification cluster

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

2.1K
Combining Reflectance Confocal Microscopy with Optical Coherence Tomography for Noninvasive Diagnosis of Skin Cancers via Image Acquisition
09:37

Combining Reflectance Confocal Microscopy with Optical Coherence Tomography for Noninvasive Diagnosis of Skin Cancers via Image Acquisition

Published on: August 18, 2022

2.6K

Related Experiment Videos

Last Updated: Oct 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

43.1K
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

2.1K
Combining Reflectance Confocal Microscopy with Optical Coherence Tomography for Noninvasive Diagnosis of Skin Cancers via Image Acquisition
09:37

Combining Reflectance Confocal Microscopy with Optical Coherence Tomography for Noninvasive Diagnosis of Skin Cancers via Image Acquisition

Published on: August 18, 2022

2.6K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Oncology

Background:

  • Classifying benign and malignant microcalcification clusters (MCs) is crucial for computer-aided diagnosis (CAD) in digital breast tomosynthesis (DBT).
  • DBT's anisotropic resolution and varying MC sharpness across slices pose challenges for standard 3D convolutional neural networks (CNNs).
  • Existing CAD algorithms struggle with the unique characteristics of DBT imaging, limiting diagnostic accuracy.

Purpose of the Study:

  • To develop an ensemble CNN that leverages both 2D focus slice and 3D contextual features from DBT images.
  • To address the anisotropic resolution inherent in DBT imaging for improved MC classification.
  • To enhance the performance of CAD systems for detecting benign versus malignant MCs.

Main Methods:

  • Proposed a novel ensemble CNN combining 2D ResNet34 for focus slice features and an anisotropic 3D ResNet for 3D contextual features.
  • Utilized anisotropic 3D convolution to mitigate the impact of DBT's anisotropic resolution.
  • Evaluated the model on 495 MCs from 275 patients' DBT images.

Main Results:

  • The ensemble CNN achieved an Area Under the Curve (AUC) of 0.8837 and an accuracy of 82.00% for MC classification.
  • Performance significantly surpassed individual 2D ResNet34 (AUC: 0.8264, ACC: 76.00%) and anisotropic 3D ResNet (AUC: 0.8455, ACC: 76.00%).
  • Demonstrated improvements in AUC (0.0435), F1 score (79.37 to 85.71%), sensitivity (78.13 to 84.38%), and specificity (66.67 to 77.78%) compared to radiomics, effectively reducing false positives.

Conclusions:

  • The ensemble CNN effectively integrates 2D and 3D features for superior classification of MCs in DBT images.
  • The proposed method significantly improves diagnostic performance and reduces false positives in breast cancer screening.
  • This approach offers a promising advancement for computer-aided diagnosis in DBT.