Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

IPFL : Interpretable Federated Learning for Personalized Healthcare

Nijdam, Arthur A. LU and Aminifar, Amir LU orcid (2025) In IEEE Access 13. p.171156-171169
Abstract

Federated Learning (FL) enables decentralized training of neural networks across multiple hospitals or patients while preserving data privacy. However, FL schemes typically assume data is independent and identically distributed (IID) while healthcare data can be highly heterogeneous. To address this, we propose Interpretable Personalized Federated Learning (IPFL), a novel framework that allows patients to selectively collaborate with others based on both validation performance and historical collaboration success. By directly inferring patient similarities from data, IPFL enables personalized model training without requiring assumptions about cluster structures or interpolation with a global model. We validate IPFL on two real-world... (More)

Federated Learning (FL) enables decentralized training of neural networks across multiple hospitals or patients while preserving data privacy. However, FL schemes typically assume data is independent and identically distributed (IID) while healthcare data can be highly heterogeneous. To address this, we propose Interpretable Personalized Federated Learning (IPFL), a novel framework that allows patients to selectively collaborate with others based on both validation performance and historical collaboration success. By directly inferring patient similarities from data, IPFL enables personalized model training without requiring assumptions about cluster structures or interpolation with a global model. We validate IPFL on two real-world healthcare tasks: epileptic seizure detection and cardiac arrhythmia detection, and show that it achieves state-of-the-art performance. Moreover, our analysis demonstrates that IPFL naturally leads to interpretable collaboration graphs: patients with similar disease characteristics tend to collaborate more frequently.

(Less)
Please use this url to cite or link to this publication:
author
and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
arrhythmia, clustering methods, epilepsy, Federated learning, Internet of Things, machine learning, precision medicine
in
IEEE Access
volume
13
pages
14 pages
publisher
IEEE - Institute of Electrical and Electronics Engineers Inc.
external identifiers
  • scopus:105015887372
ISSN
2169-3536
DOI
10.1109/ACCESS.2025.3608852
language
English
LU publication?
yes
id
38ba8542-a09a-4df5-9624-68a38ef0d200
date added to LUP
2025-11-12 12:23:00
date last changed
2025-11-12 12:23:57
@article{38ba8542-a09a-4df5-9624-68a38ef0d200,
  abstract     = {{<p>Federated Learning (FL) enables decentralized training of neural networks across multiple hospitals or patients while preserving data privacy. However, FL schemes typically assume data is independent and identically distributed (IID) while healthcare data can be highly heterogeneous. To address this, we propose Interpretable Personalized Federated Learning (IPFL), a novel framework that allows patients to selectively collaborate with others based on both validation performance and historical collaboration success. By directly inferring patient similarities from data, IPFL enables personalized model training without requiring assumptions about cluster structures or interpolation with a global model. We validate IPFL on two real-world healthcare tasks: epileptic seizure detection and cardiac arrhythmia detection, and show that it achieves state-of-the-art performance. Moreover, our analysis demonstrates that IPFL naturally leads to interpretable collaboration graphs: patients with similar disease characteristics tend to collaborate more frequently.</p>}},
  author       = {{Nijdam, Arthur A. and Aminifar, Amir}},
  issn         = {{2169-3536}},
  keywords     = {{arrhythmia; clustering methods; epilepsy; Federated learning; Internet of Things; machine learning; precision medicine}},
  language     = {{eng}},
  pages        = {{171156--171169}},
  publisher    = {{IEEE - Institute of Electrical and Electronics Engineers Inc.}},
  series       = {{IEEE Access}},
  title        = {{IPFL : Interpretable Federated Learning for Personalized Healthcare}},
  url          = {{http://dx.doi.org/10.1109/ACCESS.2025.3608852}},
  doi          = {{10.1109/ACCESS.2025.3608852}},
  volume       = {{13}},
  year         = {{2025}},
}