What does this research mean for the field?

The FedTM framework improves Parkinson’s disease classification performance, increasing AUROC from 77% to 82%, while reducing misclassification disparities and training load for smaller sites. Novelty: ClaimNovelty.NOVEL_FINDING. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

This research aims to enhance model training in digital health while preserving patient privacy, focusing on performance improvement and equity.

March 1, 2026Open Access

Combining federated learning and travelling model boosts performance and opens opportunities for digital health equity

Puntos clave

This research aims to enhance model training in digital health while preserving patient privacy, focusing on performance improvement and equity.
Developed a hybrid framework, FedTM, integrating federated learning and travelling model approaches.
Conducted warmup training using federated learning at sites with larger datasets.
Implemented sequential refinement through the travelling model across all participating sites.
Evaluated performance using 1817 brain scans for Parkinson's disease from 83 international sites.
Achieved an improvement in AUROC from 77 ± 0.01% to 82 ± 0.01%.
Reduced misclassification disparities from 34 ± 0.01% to 26 ± 0.01%.
Decreased the training load for smaller sites from 22 cycles to 12 cycles.

Resumen

Federated learning (FL) and travelling model (TM) allow privacy-preserving model training across sites without sharing patient-sensitive data. While both approaches have shown success, they face unique challenges related to distribution shifts between sites. To address this, we propose FedTM, a hybrid framework combining the strengths of FL and TM. FedTM begins with FL warmup training at sites with larger datasets, followed by sequential refinement through TM across all sites. We evaluated FedTM for Parkinson’s disease classification using 1817 brain scans from 83 international sites. Model performance, misclassification disparities, and communication costs were computed and compared to standard FL and TM approaches. Our results reveal that FedTM improves AUROC from 77 ± 0.01% to 82 ± 0.01%, reduces misclassification disparities from 34 ± 0.01% to 26 ± 0.01%, and decreases training load for smaller sites from 22 to 12 cycles. These advancements mark an important step toward promoting global healthcare equity and advancing responsible AI development.

Me gusta

Guardar

Ver artículo completo