Contrastive prototype federated learning against noisy labels in fetal standard plane detection
International Journal of Computer Assisted Radiology and Surgery · 21.05.2025 · 20 (7) · 1431-1439
DOI: 10.1007/s11548-025-03400-6 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Computer Assisted Radiology and Surgery, online erschienen am 21.05.2025. Verfasst von Maria Chiara Fiorentino und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Information Engineering, Università Politecnica delle Marche, Ancona, Italy. m.c.fiorentino@staff.univpm.it. Hauptthemen laut MEDLINE: Ultrasonography, Prenatal, Machine Learning, Image Interpretation, Computer-Assisted. Die Arbeit wurde bisher 14-mal zitiert (Crossref) und nennt 22 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Università Politecnica delle Marche.
Abstract (Originalsprache)
Purpose
This study aims to improve federated learning (FL) for ultrasound fetal standard plane detection by addressing noisy labels and data size variability across decentralized clients. We propose a federated denoising framework leveraging prototypes from the largest dataset in the federation to refine noisy labels and enhance predictions in all clients while preserving privacy.
Methods
The proposed framework consists of two main steps. First, contrastive learning (SimCLR) is applied to the images of the largest client, generating robust embeddings. These embeddings are used to refine noisy labels in the same client by leveraging the latent space structure using a threshold-based k-nearest neighbors re-labeling strategy. As a second step, image prototypes, computed from the embeddings with noise-free labels, along with SimCLR trained backbone, are shared with the smallest client to guide the FL process effectively, without requiring the use of labels from the smallest client. To address possible image distribution shifts, an ensemble strategy is introduced, which uses a majority voting scheme to optimize label refinement in the smallest dataset while minimizing image discard.
Results
Our framework showed improved performance compared to traditional FL approaches in standard plane detection, achieving the highest mean F1-score across planes.
Conclusions
The proposed strategy effectively improves fetal standard plane detection by leveraging high-quality prototypes, enabling robust performance even with noisy and heterogeneous data size across clients, while preserving privacy.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Information Engineering, Università Politecnica delle Marche, Ancona, Italy. m.c.fiorentino@staff.univpm.it.
- Department of Law, Università degli Studi di Macerata, Macerata, Italy.
- Department of Information Engineering, Università Politecnica delle Marche, Ancona, Italy.
- Department of Political Sciences, Communication and International Relations, Università degli Studi di Macerata, Macerata, Italy.
- Department of Innovative Technologies in Medicine and Dentistry, Università degli Studi "G. d'Annunzio", Chieti, Pescara, Italy.
Förderung
- Università Politecnica delle Marche
Zitieren
APA
Fiorentino, M. C., Migliorelli, G., Villani, F. P., Frontoni, E., & Moccia, S. (2025). Contrastive prototype federated learning against noisy labels in fetal standard plane detection. International Journal of Computer Assisted Radiology and Surgery, 20(7), 1431-1439. https://doi.org/10.1007/s11548-025-03400-6
Vancouver
Fiorentino MC, Migliorelli G, Villani FP, Frontoni E, Moccia S. Contrastive prototype federated learning against noisy labels in fetal standard plane detection. International Journal of Computer Assisted Radiology and Surgery. 2025;20(7):1431-1439. doi:10.1007/s11548-025-03400-6
BibTeX
@article{Fiorentino2025,
author = {Fiorentino, Maria Chiara and Migliorelli, Giovanna and Villani, Francesca Pia and Frontoni, Emanuele and Moccia, Sara},
title = {Contrastive prototype federated learning against noisy labels in fetal standard plane detection},
journal = {International Journal of Computer Assisted Radiology and Surgery},
year = {2025},
volume = {20},
number = {7},
pages = {1431--1439},
doi = {10.1007/s11548-025-03400-6}
}
EUVIMED