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

WDK-Net: Lightweight Wavelet Diffusion with Kolmogorov-Arnold Network for Limited-angle Cardiac CT Reconstruction.

IEEE transactions on medical imaging·2026
Same author

Clinical Metadata-Guided Limited-Angle CT Image Reconstruction.

IEEE transactions on medical imaging·2026
Same author

Gradient Guided Co-Retention Feature Pyramid Network for LDCT Image Denoising.

Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention·2025
Same author

Clinical Metadata Guided Limited-Angle CT Image Reconstruction.

ArXiv·2025
Same author

Leveraging an ensemble of EfficientNetV1 and EfficientNetV2 models for classification and interpretation of breast cancer histopathology images.

Scientific reports·2025
Same author

A comparative study of rigid-body registration algorithms for the alignment of longitudinal structural MRI of the brain.

Journal of neuroscience methods·2025

Related Experiment Video

Updated: Jul 20, 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

42.9K

Reduced Deep Convolutional Activation Features (R-DeCAF) in Histopathology Images to Improve the Classification

Bahareh Morovati1, Reza Lashgari1, Mojtaba Hajihasani1

  • 1Institute of Medical Science and Technology, Shahid Beheshti University, Tehran, Iran.

Journal of Digital Imaging
|August 2, 2023
PubMed
Summary

This study introduces reduced deep convolutional activation features (R-DeCAF) to improve breast cancer classification accuracy. By applying dimension reduction techniques like PCA, R-DeCAF enhances diagnostic efficiency for pathologists.

Keywords:
Breast cancerDeep feature extractionFeature reductionHistopathology imagesPre-trained convolutional neural networks

More Related Videos

Detection and Isolation of Cancer in Prostate Biopsies Using Stimulated Raman Histology and Artificial Intelligence
08:08

Detection and Isolation of Cancer in Prostate Biopsies Using Stimulated Raman Histology and Artificial Intelligence

Published on: June 10, 2025

93
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

8.3K

Related Experiment Videos

Last Updated: Jul 20, 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

42.9K
Detection and Isolation of Cancer in Prostate Biopsies Using Stimulated Raman Histology and Artificial Intelligence
08:08

Detection and Isolation of Cancer in Prostate Biopsies Using Stimulated Raman Histology and Artificial Intelligence

Published on: June 10, 2025

93
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

8.3K

Area of Science:

  • Medical Imaging and Diagnostics
  • Artificial Intelligence in Healthcare
  • Computational Pathology

Background:

  • Breast cancer diagnosis by pathologists is time-consuming and subjective.
  • Computer-aided diagnosis (CAD) systems, particularly those using deep convolutional neural networks (CNNs), offer automated classification to aid pathologists.
  • Deep convolutional activation features (DeCAF) extracted from pre-trained CNNs are valuable but require optimization.

Purpose of the Study:

  • To investigate the role of dimension reduction in optimizing DeCAF for breast cancer classification.
  • To propose a reduced DeCAF (R-DeCAF) framework to enhance classification accuracy and efficiency.
  • To identify effective dimension reduction methods for combining deep features.

Main Methods:

  • Utilized pre-trained CNNs (AlexNet, VGG-16, VGG-19) in transfer learning mode for feature extraction.
  • Extracted DeCAF from the first fully connected layer.
  • Applied various linear and nonlinear dimension reduction techniques, focusing on Principal Component Analysis (PCA), and used a Support Vector Machine (SVM) for classification.

Main Results:

  • Analysis revealed that not all DeCAF features contribute equally to accuracy, highlighting the importance of dimension reduction.
  • Linear dimension reduction, specifically PCA, demonstrated a superior ability to combine deep features effectively.
  • The proposed R-DeCAF method achieved up to a 4.3% improvement in classification accuracy with a feature vector size (FVS) of 23 and cumulative explained variance (CEV) of 0.15.

Conclusions:

  • Dimension reduction plays a critical role in enhancing the performance of DeCAF for breast cancer classification.
  • The R-DeCAF framework, particularly using PCA, offers a more efficient and accurate approach to computer-aided breast cancer diagnosis.
  • The findings suggest that optimized feature selection can significantly improve diagnostic accuracy in computational pathology.