Federated Learning and Differential Privacy Techniques on Multi-hospital Population-scale Electrocardiogram Data
Authors
Vikhyat Agrawal and Sunil Vasu Kalmady and Venkataseetharam Manoj Malipeddi and Manisimha Varma Manthena and Weijie Sun and Saiful Islam and Abram Hindle and Padma Kaul and Russell Greiner
Venue
- International Conference on Medical and Health Informatics (ICMHI 2024)
- Yokohama, Japan
- 2024
- 1–9
Abstract
This research paper explores ways to apply Federated Learning (FL) and Differential Privacy (DP) techniques to population-scale Electrocardiogram (ECG) data. The study learns a multi-label ECG classification model using FL and DP based on 1,565,849 ECG tracings from 7 hospitals in Alberta, Canada. The FL approach allowed collaborative model training without sharing raw data between hospitals while building robust ECG classification models for diagnosing various cardiac conditions. These accurate ECG classification models can facilitate the diagnoses while preserving patient confidentiality using FL and DP techniques. Our results show that the performance achieved using our implementation of the FL approach is comparable to that of the pooled approach, where the model is trained over the aggregating data from all hospitals. Furthermore, our findings suggest that hospitals with limited ECGs for training can benefit from adopting the FL model compared to single-site training. In addition, this study showcases the trade-off between model performance and data privacy by employing DP during model training.
Bibtex
@inproceedings{agrawal2024ICMHI-federated,
abstract = {This research paper explores ways to apply Federated Learning (FL) and Differential Privacy (DP) techniques to population-scale Electrocardiogram (ECG) data. The study learns a multi-label ECG classification model using FL and DP based on 1,565,849 ECG tracings from 7 hospitals in Alberta, Canada. The FL approach allowed collaborative model training without sharing raw data between hospitals while building robust ECG classification models for diagnosing various cardiac conditions. These accurate ECG classification models can facilitate the diagnoses while preserving patient confidentiality using FL and DP techniques. Our results show that the performance achieved using our implementation of the FL approach is comparable to that of the pooled approach, where the model is trained over the aggregating data from all hospitals. Furthermore, our findings suggest that hospitals with limited ECGs for training can benefit from adopting the FL model compared to single-site training. In addition, this study showcases the trade-off between model performance and data privacy by employing DP during model training.},
accepted = {2024-02-04},
author = {Vikhyat Agrawal and Sunil Vasu Kalmady and Venkataseetharam Manoj Malipeddi and Manisimha Varma Manthena and Weijie Sun and Saiful Islam and Abram Hindle and Padma Kaul and Russell Greiner},
authors = {Vikhyat Agrawal and Sunil Vasu Kalmady and Venkataseetharam Manoj Malipeddi and Manisimha Varma Manthena and Weijie Sun and Saiful Islam and Abram Hindle and Padma Kaul and Russell Greiner},
booktitle = {International Conference on Medical and Health Informatics (ICMHI 2024)},
code = {agrawal2024ICMHI-federated},
date = {2024-05-15},
funding = {NSERC Discovery},
location = {Yokohama, Japan},
pages = {1--9},
role = {Editorial},
title = {Federated Learning and Differential Privacy Techniques on Multi-hospital Population-scale Electrocardiogram Data},
type = {inproceedings},
url = {http://softwareprocess.ca/pubs/agrawal2024ICMHI-federated.pdf},
venue = {International Conference on Medical and Health Informatics (ICMHI 2024)},
year = {2024}
}