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

Thrombin Concentration Shapes Endothelial Extracellular Vesicle Profiles with Divergent Inflammatory Functions.

Blood advances·2026
Same author

Activated protein C promotes human lung cancer progression through the release of tumor extracellular vesicles and transfer of microRNA-200a.

Cell death & disease·2025
Same author

Implantation in single euploid embryo transfers: the role of blastocyst expansion grade and early serum <i>β</i>-hCG levels.

Gynecological endocrinology : the official journal of the International Society of Gynecological Endocrinology·2025
Same author

High-dimensional item response theory analysis of patient-reported outcomes in total knee arthroplasty.

NPJ digital medicine·2025
Same author

Assisted Reproduction Technology Treatment Outcomes in Female Carriers of 21-Hydroxylase Deficiency.

The Journal of clinical endocrinology and metabolism·2025
Same author

Development and multicentric external validation of a prognostic COVID-19 severity model based on thoracic CT.

BMC medical informatics and decision making·2025

Related Experiment Video

Updated: Jul 2, 2025

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

2.8K

Semi-supervised medical image classification via distance correlation minimization and graph attention

Abel Díaz Berenguer1, Maryna Kvasnytsia1, Matías Nicolás Bossa1

  • 1Vrije Universiteit Brussel (VUB), Department of Electronics and Informatics (ETRO), Pleinlaan 2, 1050 Brussels, Belgium.

Medical Image Analysis
|February 24, 2024
PubMed
Summary
This summary is machine-generated.

This study introduces a new semi-supervised learning method for medical imaging classification, effectively using unlabeled data to boost performance with limited annotations. The approach enhances accuracy by minimizing feature correlations and modeling image relationships.

Keywords:
Deep neural networksDistance correlationGraph attentionMedical image classificationSemi-supervised learning

More Related Videos

Author Spotlight: Advancing Cardiovascular Imaging - Introducing the Spatially Weighted Calcium Score for Early Disease Detection
06:57

Author Spotlight: Advancing Cardiovascular Imaging - Introducing the Spatially Weighted Calcium Score for Early Disease Detection

Published on: September 22, 2023

985
Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies
04:25

Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies

Published on: December 15, 2023

2.4K

Related Experiment Videos

Last Updated: Jul 2, 2025

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

2.8K
Author Spotlight: Advancing Cardiovascular Imaging - Introducing the Spatially Weighted Calcium Score for Early Disease Detection
06:57

Author Spotlight: Advancing Cardiovascular Imaging - Introducing the Spatially Weighted Calcium Score for Early Disease Detection

Published on: September 22, 2023

985
Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies
04:25

Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies

Published on: December 15, 2023

2.4K

Area of Science:

  • Medical Imaging Analysis
  • Machine Learning
  • Computer Vision

Background:

  • Medical imaging classification relies heavily on annotated data, which is often scarce and expensive to obtain.
  • Limited labeled datasets hinder the performance of deep learning models in realistic medical scenarios.
  • Developing cost-effective methods to leverage abundant unlabeled medical data is crucial.

Purpose of the Study:

  • To propose a novel semi-supervised learning method for medical imaging classification that utilizes unlabeled data alongside minimal annotated data.
  • To improve classification performance in resource-constrained settings with limited data annotation budgets.
  • To introduce and validate new techniques for feature representation and data regularization in medical image analysis.

Main Methods:

  • A novel semi-supervised learning approach incorporating distance correlation to minimize feature representation correlations between different image views.
  • Utilizing non-coupled deep neural network architectures for feature encoding.
  • Implementing a data-driven graph-attention based regularization strategy to model affinities within unlabeled data based on feature-space relationships.

Main Results:

  • The proposed method achieved highly competitive performance across four diverse medical imaging datasets (X-ray, dermoscopic, MRI, CT).
  • The approach demonstrated superior performance compared to several state-of-the-art semi-supervised learning methods.
  • Experiments validated the effectiveness of distance correlation and graph-attention regularization for medical image analysis.

Conclusions:

  • The novel semi-supervised learning method significantly enhances medical imaging classification accuracy with limited annotations.
  • Distance correlation proves to be a versatile measure for reducing feature dependencies.
  • Graph-attention based regularization effectively models image affinities, benefiting semi-supervised learning in medical imaging.