EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Advancing remote ultrasound diagnosis through a federated deep transfer learning framework

Oluwatobi Noah Akande, Funmilola Alaba Ajala, Gift Adene, Julius Makinde, Roseline Oluwaseun Ogundokun, Rotimi-Williams Bello

Journal of the Nigerian Society of Physical Sciences · 17.11.2026 · 3779

DOI: 10.46481/jnsps.2027.3779 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of the Nigerian Society of Physical Sciences, online erschienen am 17.11.2026. Verfasst von Oluwatobi Noah Akande und 5 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Telerobotic ultrasound systems enable clinicians to remotely control robotic arms equipped with ultrasound probes, improving diagnostic access. However, their effectiveness is constrained by limited annotated datasets, privacy concerns associated with centralized storage, and a lack of transparent and generalizable artificial intelligence (AI) models. This study presents a Federated Deep Transfer Learning (FDTL) framework supporting collaborative ultrasound image classification while keeping images on local client nodes. Four pretrained convolutional neural networks (CNNs) were fine-tuned on the breast ultrasound images (BUSI) dataset to extract diagnostic features. Federated averaging (FedAvg) enabled model aggregation without sharing raw client data. Centralized and federated approaches were compared, and convergence was analyzed. A gradient-weighted class activation mapping (Grad-CAM) module supported interpretability. The framework achieved a peak federated classification accuracy of 91.1% using DenseNet-121 and a macro-averaged accuracy of 90.5% across the evaluated architectures, outperforming individual client models while maintaining stable convergence over 80 communication rounds. Convergence stabilized within 50--60 rounds. Grad-CAM highlighted attention regions corresponding to diagnostically relevant lesion boundaries and textures. The framework shows a promising balance of diagnostic performance, data localization, interpretability, and scalability under simulated federated conditions. These findings provide preliminary evidence of robustness across heterogeneous simulated clients and suggest potential for decentralized ultrasound diagnosis.

Zitieren

APA

Akande, O. N., Ajala, F. A., Adene, G., Makinde, J., Ogundokun, R. O., & Bello, R. (2026). Advancing remote ultrasound diagnosis through a federated deep transfer learning framework. Journal of the Nigerian Society of Physical Sciences, 3779. https://doi.org/10.46481/jnsps.2027.3779

Vancouver

Akande ON, Ajala FA, Adene G, Makinde J, Ogundokun RO, Bello R. Advancing remote ultrasound diagnosis through a federated deep transfer learning framework. Journal of the Nigerian Society of Physical Sciences. 2026:3779. doi:10.46481/jnsps.2027.3779

BibTeX
@article{Akande2026,
  author = {Akande, Oluwatobi Noah and Ajala, Funmilola Alaba and Adene, Gift and Makinde, Julius and Ogundokun, Roseline Oluwaseun and Bello, Rotimi-Williams},
  title = {Advancing remote ultrasound diagnosis through a federated deep transfer learning framework},
  journal = {Journal of the Nigerian Society of Physical Sciences},
  year = {2026},
  pages = {3779},
  doi = {10.46481/jnsps.2027.3779}
}
Kostenlos starten →