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: Oct 5, 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

3.0K

Semi-supervised learning for medical image classification using imbalanced training data.

Tri Huynh1, Aiden Nibali1, Zhen He1

  • 1Department of Computer Science and Information Technology, La Trobe University, Melbourne, Australia.

Computer Methods and Programs in Biomedicine
|February 1, 2022
PubMed
Summary
This summary is machine-generated.

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

Development of a serverless, interactive application for Alzheimer's disease detection and visualization using MRI images.

The neuroradiology journal·2026
Same author

'We want beer': an analysis of online posts written about the alcohol ban during Qatar's soccer world cup.

Alcohol and alcoholism (Oxford, Oxfordshire)·2025
Same author

OCELOT 2023: Cell detection from cell-tissue interaction challenge.

Medical image analysis·2025
Same author

Patient-Reported Favorable Outcomes of a Minimum Five-Year Follow-Up After Medial Patellofemoral Ligament (MPFL) Reconstruction: A Systematic Review and Meta-Analysis.

Cureus·2025
Same author

Automated deep learning-based assessment of tumour-infiltrating lymphocyte density determines prognosis in colorectal cancer.

Journal of translational medicine·2025
Same author

Shedding light on the role of complement C4 activation in cancer.

Human immunology·2024
Same journal

Facial iPPG heatmap patterns based on period-aware autoencoder show association with carotid atherosclerosis towards non-contact hemodynamic assessment.

Computer methods and programs in biomedicine·2026
Same journal

Explainable machine learning models predict liver fibrosis risk and outcome in the general population: Development and multi-cohort external validation.

Computer methods and programs in biomedicine·2026
Same journal

Evaluation of surrogate endpoints for survival outcomes using the surrogate package in R.

Computer methods and programs in biomedicine·2026
Same journal

Relative spectral and frication-based descriptors as numerical indicators of place of articulation shifts in fricatives produced by Polish children.

Computer methods and programs in biomedicine·2026
Same journal

Leaflet resection improves valve expansion and hemodynamic performance in redo TAVI with balloon- and self-expanding transcatheter heart valve configurations.

Computer methods and programs in biomedicine·2026
Same journal

Spectral super-resolution for Parkinson's voice via representation-level methods under mixed-reality acquisition.

Computer methods and programs in biomedicine·2026
See all related articles

This study introduces Adaptive Blended Consistency Loss (ABCL), a new method for semi-supervised learning in medical image classification. ABCL effectively addresses class imbalance, improving diagnostic accuracy for rare diseases.

Area of Science:

  • Medical image analysis
  • Machine learning for healthcare
  • Computer-aided diagnosis

Background:

  • Medical image classification faces challenges due to limited labeled data and imbalanced class distributions.
  • Existing semi-supervised learning methods often fail to address class imbalance effectively.
  • This necessitates novel approaches for accurate medical image classification with skewed datasets.

Purpose of the Study:

  • To introduce a new perturbation-based semi-supervised learning approach for medical image classification.
  • To specifically tackle the challenge of imbalanced training data in medical image datasets.
  • To improve the performance of semi-supervised learning models on underrepresented disease classes.

Main Methods:

  • Propose Adaptive Blended Consistency Loss (ABCL) as a drop-in replacement for consistency loss in semi-supervised learning.
Keywords:
Class imbalanceMedical imagingSemi supervised learning

Related Experiment Videos

Last Updated: Oct 5, 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

3.0K
  • ABCL adaptively adjusts the target class distribution of consistency loss based on class frequency to counteract data skew.
  • Evaluate ABCL on imbalanced skin cancer and retinal fundus glaucoma datasets, comparing it with existing methods and performing an ablation study.

    • Main Results:

    • ABCL significantly improves unweighted average recall (UAR) compared to existing consistency losses.
    • On the skin cancer dataset, ABCL boosted performance from 0.59 to 0.67 UAR, outperforming other imbalance-addressing methods.
    • On the glaucoma dataset, ABCL achieved 0.67 UAR, surpassing the best existing approach (0.57 UAR).

    Conclusions:

    • Adaptive Blended Consistency Loss (ABCL) effectively alleviates class imbalance in semi-supervised medical image classification.
    • The proposed method offers a practical solution for improving diagnostic accuracy in medical imaging.
    • ABCL demonstrates significant potential for enhancing machine learning applications in healthcare where data imbalance is prevalent.