A decentralized framework for biostatistics and privacy concerns

Mangold, Paul; Moussa, Mouhamed; Sobanski, Vincent; Ficheur, Gregoire; Andrey, Paul; Lamer, Antoine

Document type :

Article dans une revue scientifique: Article original

DOI :

10.3233/SHTI200710

PMID :

33227756

Permalink :

http://hdl.handle.net/20.500.12210/40619

Title :

A decentralized framework for biostatistics and privacy concerns

Author(s) :

Mangold, Paul [Auteur]
Machine Learning in Information Networks [MAGNET]
Evaluation des technologies de santé et des pratiques médicales - ULR 2694 [METRICS]
Centre Hospitalier Régional Universitaire [CHU Lille] [CHRU Lille]
École normale supérieure de Lyon [ENS de Lyon]
Moussa, Mouhamed [Auteur]

Récepteurs nucléaires, Maladies Cardiovasculaires et Diabète (EGID) - U1011
Centre Hospitalier Régional Universitaire [CHU Lille] [CHRU Lille]
Récepteurs nucléaires, Maladies Cardiovasculaires et Diabète (EGID) - U1011
Sobanski, Vincent [Auteur]

Centre Hospitalier Régional Universitaire [CHU Lille] [CHRU Lille]
Ficheur, Gregoire [Auteur]

METRICS : Evaluation des technologies de santé et des pratiques médicales - ULR 2694
Centre Hospitalier Régional Universitaire [CHU Lille] [CHRU Lille]
Andrey, Paul [Auteur]
Centre Hospitalier Régional Universitaire [CHU Lille] [CHRU Lille]
Lamer, Antoine [Auteur]

METRICS : Evaluation des technologies de santé et des pratiques médicales - ULR 2694

Journal title :

Studies in Health Technology and Informatics

Abbreviated title :

Stud Health Technol Inform

Volume number :

275

Pages :

137-141

Publication date :

2020-11-23

ISSN :

1879-8365

HAL domain(s) :

Sciences du Vivant [q-bio]

English abstract : [en]

Biostatistics and machine learning have been the cornerstone of a variety of recent developments in medicine. In order to gather large enough datasets, it is often necessary to set up multi-centric studies; yet, centralization ...
Show more >Biostatistics and machine learning have been the cornerstone of a variety of recent developments in medicine. In order to gather large enough datasets, it is often necessary to set up multi-centric studies; yet, centralization of measurements can be difficult, either for practical, legal or ethical reasons. As an alternative, federated learning enables leveraging multiple centers' data without actually collating them. While existing works generally require a center to act as a leader and coordinate computations, we propose a fully decentralized framework where each center plays the same role. In this paper, we apply this framework to logistic regression, including confidence intervals computation. We test our algorithm on two distinct clinical datasets split among different centers, and show that it matches results from the centralized framework. In addition, we discuss possible privacy leaks and potential protection mechanisms, paving the way towards further research.Show less >

Language :

Anglais

Audience :

Internationale

Popular science :

Non

Administrative institution(s) :

CHU Lille
Inserm
Institut Pasteur de Lille
Université de Lille

Collections :

Submission date :

2021-07-06T12:45:59Z
2024-01-29T09:05:20Z

Files

SHTI-275-SHTI200710.pdf
Non spécifié
Open access
Access the document

Except where otherwise noted, this item's license is described as Attribution-NonCommercial 3.0 United States

Version	Link	Modification date
2	20.500.12210/40619*	2024-01-29T08:40:47Z
1	20.500.12210/40619.1	2021-07-06T12:45:59Z

A decentralized framework for biostatistics ...

BibTeX

CSV

Excel

RIS

Files

Version history

A decentralized framework for biostatistics ... BibTeX CSV Excel RIS

Files

Version history

A decentralized framework for biostatistics ...

BibTeX

CSV

Excel

RIS