EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Federated Learning for Multi-Disease Ophthalmic Diagnostics Using OCT Angiography

Ahammed Sakir Nabil, Sina Gholami, Theodore Leng, Jennifer I Lim, Minhaj Nur Alam

Ophthalmology Science · 2026-03 · 6 (3) · 101030

DOI: 10.1016/j.xops.2025.101030 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Ophthalmology Science, 2026-03. Verfasst von Ahammed Sakir Nabil und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Electrical and Computer Engineering, University of North Carolina at Charlotte, Charlotte, North Carolina. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 67 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: NIDDK NIH HHS, NEI NIH HHS, National Eye Institute.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Purpose

To conduct a comprehensive systematic evaluation of federated learning (FL) strategies for multi-disease retinal classification using OCT angiography (OCTA), implementing a 2-part experimental framework to establish foundational feasibility and optimize performance under realistic heterogeneous conditions while ensuring privacy preservation.

Design

Retrospective multi-center FL study using a systematic 2-part experimental design: (1) foundational feasibility evaluation under controlled homogeneous conditions, and (2) comprehensive optimization under realistic heterogeneous conditions using Dirichlet distribution partitioning (α = 0.5).

Participants

A total of 456 OCTA images from patients with 7 retinal pathologies, with diabetic retinopathy (31.1%) and normal cases (25.2%) comprising the majority, sourced from the public OCTA-500 data set (n = 300) and a private collection from the University of Illinois Chicago (n = 156).

Methods

Five FL aggregation strategies (federated averaging [FedAvg], federated proximal [FedProx], federated magnetic resonance imaging [FedMRI], federated Adagrad, and federated Yogi) were systematically evaluated across multiple optimization dimensions: 7 architecture configurations spanning vision transformers, established convolutional neural networks, and hybrid models; 5 transfer learning freezing strategies; 3 local epoch configurations (2, 5, and 10); and scalability analysis across 2, 3, and 5-client federations. Security mechanisms including differential privacy (ε = 1.0-8.0) and secure aggregation were integrated and evaluated. Performance was assessed across 3 classification scenarios: 7-class, 4-class modified, and 4-class streamlined. Main Outcome Measures: Classification accuracy, receiver-operating-characteristic area under the curve (ROC-AUC), and macro-averaged F1-score with comprehensive privacy-utility analysis and computational efficiency metrics.

Results

Under controlled conditions, FL achieved superior performance in simplified classifications, with FedAvg, FedProx, and FedMRI reaching 72.09% accuracy versus 69.77% centralized training. Comprehensive optimization identified DenseNet121 as optimal architecture (79.55% accuracy, 89.68% ROC-AUC), with "most" freezing strategy (75% frozen layers) providing 60% training time reduction while maintaining superior performance. Federated proximal demonstrated exceptional resilience to heterogeneity (-11.7% degradation). Bonawitz secure aggregation achieved optimal privacy-utility balance (63.64% accuracy with cryptographic guarantees), whereas differential privacy maintained clinical utility under moderate constraints (ε ≈ 4-6).

Conclusions

This systematic evaluation establishes FL as a comprehensive solution for privacy-preserving multi-institutional OCTA-based disease classification, with careful architectural selection, optimization strategies, and security mechanisms enabling performance that matches or exceeds centralized approaches while maintaining regulatory compliance and clinical utility. Financial Disclosures: The authors have no proprietary or commercial interest in any materials discussed in this article.

Schlagwörter

Stichwörter der Autor:innen

Age related macular degeneration (AMD)Diabetic retinopathy (DR)Federated learningOptical coherence tomography angiography (OCTA)Privacy-preserving artificial intelligence

Autor:innen und Institutionen

  • Department of Electrical and Computer Engineering, University of North Carolina at Charlotte, Charlotte, North Carolina.
  • Department of Ophthalmology, Byers Eye Institute at Stanford, Stanford University School of Medicine, Stanford, California.
  • Department of Ophthalmology and Visual Science, University of Illinois Chicago, Chicago, Illinois.

Förderung

  • NIDDK NIH HHS
  • NEI NIH HHS
  • National Eye Institute
  • National Institutes of Health
  • Research to Prevent Blindness
  • University of North Carolina
  • UNC Charlotte

Zitieren

APA

Nabil, A. S., Gholami, S., Leng, T., Lim, J. I., & Alam, M. N. (2026). Federated Learning for Multi-Disease Ophthalmic Diagnostics Using OCT Angiography. Ophthalmology Science, 6(3), 101030. https://doi.org/10.1016/j.xops.2025.101030

Vancouver

Nabil AS, Gholami S, Leng T, Lim JI, Alam MN. Federated Learning for Multi-Disease Ophthalmic Diagnostics Using OCT Angiography. Ophthalmology Science. 2026;6(3):101030. doi:10.1016/j.xops.2025.101030

BibTeX
@article{Nabil2026,
  author = {Nabil, Ahammed Sakir and Gholami, Sina and Leng, Theodore and Lim, Jennifer I and Alam, Minhaj Nur},
  title = {Federated Learning for Multi-Disease Ophthalmic Diagnostics Using OCT Angiography},
  journal = {Ophthalmology Science},
  year = {2026},
  volume = {6},
  number = {3},
  pages = {101030},
  doi = {10.1016/j.xops.2025.101030}
}
Kostenlos starten →