Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

A federated learning framework integrating knowledge graphs and Node2Vec for multi-source medical image

Abdullah Ali Alqarni¹

¹Department of Internal Medicine, College of Medicine, University of Bisha, Bisha, Kingdom of Saudi Arabia. almohsen@ub.edu.sa.

Scientific Reports

|June 30, 2026

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

Prevalence, side effects, and awareness about energy drink consumption among college students at a Saudi public university.

Journal of family medicine and primary care·2026

Same author

Catheter-Directed Sclerotherapy Versus Ovarian Cystectomy for Unilateral Ovarian Endometrioma: A Pilot Randomized Controlled Trial.

AJR. American journal of roentgenology·2026

Same author

The Associations of Anaemia Status and Body Mass Index with Asthma Severity in Saudi Arabia: A Comparative Study.

Journal of asthma and allergy·2025

Same author

Pituitary stalk interruption syndrome: a case report and literature review.

Annals of medicine and surgery (2012)·2024

Same author

Psoralen Isolated from the Roots of <i>Dorstenia psilurus</i> Welw. Modulate Th1/Th2 Cytokines and Inflammatory Enzymes in LPS-Stimulated RAW 264.7 Macrophages.

Mediators of inflammation·2024

Same author

Hedgehog signaling mastery: R51211's promise in augmenting the therapeutic efficacy of sorafenib.

Life sciences·2024

This study introduces GraphMedFL, a novel federated learning framework for medical image classification. It enhances accuracy and generalization by integrating knowledge graphs and structural embeddings, outperforming traditional methods.

Area of Science:

Artificial Intelligence
Medical Imaging
Machine Learning

Background:

Federated learning in healthcare faces challenges with non-IID data and limited generalization.
Conventional methods often overlook structural relationships in medical images, impacting classification performance.

Purpose of the Study:

To propose GraphMedFL, a federated learning framework enhancing medical image classification.
To improve robustness and generalization in heterogeneous healthcare environments.

Main Methods:

Constructing local knowledge graphs from image features to capture inter-sample relationships.
Integrating Node2Vec structural embeddings with feature representations for enriched local training.
Employing a similarity-aware adaptive aggregation strategy to handle non-IID data.

Keywords:

Federated learning Image classification Medical imaging Node2Vec

Related Experiment Videos

Main Results:

GraphMedFL achieved 98.3% accuracy on breast cancer datasets (BreakHis, CBIS-DDSM, INbreast).
The framework consistently outperformed baseline federated learning and local training approaches.
Demonstrated significant improvements in precision, recall, and F1-score.

Conclusions:

GraphMedFL offers a robust and generalizable solution for federated medical image classification.
The integration of knowledge graphs and structural embeddings effectively addresses data heterogeneity challenges.