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: May 17, 2026

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

GTAFL: Addressing Test-Agnostic Long-Tailed Federated Learning in Medical Image.

Guangyu Chen, Jingyun Zeng, Hui Jiang

    IEEE Journal of Biomedical and Health Informatics
    |May 15, 2026
    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

    Identification of biological raw materials in Chinese patent medicine Gejie Dingchuan Wan using high-throughput sequencing.

    Journal of ethnopharmacology·2026
    Same author

    Development of a comprehensive food phenolic acids database and its application: associations between dietary phenolic acids and risk of mortality.

    Food research international (Ottawa, Ont.)·2026
    Same author

    From access to capability: digital literacy and health inequality in contemporary China.

    Frontiers in public health·2026
    Same author

    Cross-Linked Sodium Hyaluronate Filler Containing Poly-L-Lactic Acid-b-Poly (Ethylene Glycol) Microspheres for Periorbital Rejuvenation: A Multimodal Assessment of Physicochemical, Preclinical, and Clinical Performance.

    Journal of cosmetic dermatology·2026
    Same author

    Ni<sub>5</sub>P<sub>4</sub>-CeO<sub>2</sub> heterointerface-stabilized Ru nanoparticles for efficient electrochemical water splitting.

    Journal of colloid and interface science·2026
    Same author

    Mogrosides from S. grosvenorii ameliorate acute pharyngitis via the PI3K/AKT-NF-κB-NLRP3 signaling pathway and oxidative stress modulation.

    Journal of ethnopharmacology·2026

    This study introduces Test-Agnostic Long-Tailed Federated Learning (FL) to address privacy and data challenges in healthcare AI. The proposed GTAFL framework improves model performance on unpredictable real-world medical data.

    Area of Science:

    • Artificial Intelligence
    • Machine Learning
    • Healthcare Informatics

    Background:

    • Patient privacy is paramount in healthcare due to sensitive medical data.
    • Federated Learning (FL) offers a privacy-preserving approach for decentralized AI training in healthcare.
    • Real-world medical data often presents long-tailed distributions, challenging existing FL methods.

    Purpose of the Study:

    • To address the limitations of current FL methods that assume uniform test data distributions.
    • To introduce a novel task: Test-Agnostic Long-Tailed Federated Learning.
    • To propose the GTAFL framework for robust performance on unpredictable healthcare data distributions.

    Main Methods:

    • GTAFL utilizes adaptive re-sampling, expert classifier retraining, and self-supervised learning during training.

    Related Experiment Videos

    Last Updated: May 17, 2026

    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

  • The framework corrects biased classifiers and feature spaces affected by long-tailed training data.
  • An ensemble mechanism combines retrained classifiers for inference on diverse test distributions.

    • Main Results:

    • GTAFL demonstrates superior performance compared to state-of-the-art methods.
    • Experiments were conducted on CIFAR10 and two distinct medical datasets.
    • The framework effectively handles long-tailed distributions in both training and testing phases.

    Conclusions:

    • The proposed GTAFL framework successfully addresses the challenge of test-agnostic long-tailed federated learning in healthcare.
    • GTAFL enhances the reliability and applicability of FL models in real-world, unpredictable medical data scenarios.
    • This work paves the way for more robust and privacy-preserving AI solutions in the medical domain.