Privacy Aware and Privacy Preserving Distributed and Robust Machine Learning for Medical Applications

Informations projet

PAROMA-MED

N° de convention de subvention: 101070222

DOI

10.3030/101070222

Projet clôturé

Date de signature de la CE 22 Juin 2022

Date de début 1 Juillet 2022

Date de fin 30 Juin 2025

Financé au titre de

• Civil Security for Society

Coût total

€ 4 280 497,50

Contribution de l’UE

€ 4 280 497,50

Coordonné par

EURESCOM-EUROPEAN INSTITUTE FOR RESEARCH AND STRATEGIC STUDIES IN TELECOMMUNICATIONS GMBH
Germany

CORDIS fournit des liens vers les livrables publics et les publications des projets HORIZON.

Les liens vers les livrables et les publications des projets du 7e PC, ainsi que les liens vers certains types de résultats spécifiques tels que les jeux de données et les logiciels, sont récupérés dynamiquement sur OpenAIRE .

Livrables

Documents, reports (7)

Setup of communication and dissemination media

Report about the first communication and dissemination activities, such as the public website and leaflet; it includes the project vision, objectives, activities, expected outcomes and benefits, and points of contact. Project accounts on selected social media.

1st Report and updated plan for communication, clustering, dissemination

These Deliverables will list the communication, clustering, standardization, and dissemination actions implemented every year by the Project. They will update the roadmap in D5.1 in case some targets become not feasible or do not bring the expected impact. The delivery of the three reports have been scheduling according to the expected intensity of dissemination activities (likely more concentrated in the second part of the Project, when more achievements and results will be available).

Concept and Evaluation Framework - first version

This report will provide the first version of the platform architecture and the related evaluation methodology (T1.2 and T1.3).

Security and Privacy Awareness - ver. 1

This report collects and presents all the security and privacy awareness concepts and models defined and developed in tasks T3.3. This document reports on the first iteration of the concepts and models defined and developed as well as a first prototype GUI for situational awareness.

Concept and Evaluation Framework - final version

This document will detail the updated version of the platform architecture and the related evaluation methodology as refined through the progress of the technical work performed in other WPs (T1.2 and T1.3).

Access and Privacy Control Architecture and Models

This report presents the models and the resulting architecture for access and privacy control within the PAROMA architecture. The documentation about the implementation will be part of deliverable D2.3.

Roadmap for communication, dissemination, standardization activities

This report will provide an overview of the strategy and the main actions to be implemented for communication, dissemination, standardization, clustering. It will describe relevant targets for communication/dissemination, together with the appropriate means to reach them. The roadmap will be updated yearly according to the specific achievements and relevant changes in context and objectives.

Demonstrators, pilots, prototypes (1)

Network and Interconnect Platform - ver. 1

This report collects and presents the outcomes related to the network and interconnecting platform focusing in particular about the privacy-aware Network Slices conceived, designed and implemented to assure an interconnectivity privacy level compliant with the project requirements. This report freezes the current status in the mid-lifespan of the project.

Publications

Conference proceedings (1)

Federated machine learning through edge ready architectures with privacy preservation as a service

Auteurs: Anastasius Gavras; Konstantinos Koutsopoulos; Stefan Covaci; Benjamin Ertl; Spyridon Tompros; Antoine Simon; Gouenou Coatrieux; Katarzyna Kapista
Publié dans: 2022 IEEE Future Networks World Forum (FNWF), 2023, ISBN 978-1-6654-6250-1
Éditeur: IEEE
DOI: 10.1109/FNWF55208.2022.00067

Book chapters (2)

Chapter 2: Architecture and design choices for federated learning in modern digital healthcare systems

Auteurs: Konstantinos A. Koutsopoulos, Christoph Thümmler, Angelica Avila Castillo, Alice Abend, Stefan Covaci, Benjamin Ertl, Giannis Ledakis, Stéphane Lorin, Vincent Thouvenot, Sahar Haddad, Gouenou Coatrieux, Reda Bellafqira, Alessandro Bassi
Publié dans: 2024, ISBN 9780443138966
Éditeur: Academic Press, Elsevier
DOI: 10.5281/ZENODO.12665692

Chapter 4: Recent Advances in federated learning for digital healthcare systems

Auteurs: Pooja Mohnani, Christoph Thümmler, Angelica Avila Castillo, Rasha Tolba, Alessandro Bassi, Antoine Simon, Anastasius Gavras, Orazio Toscano, Pascal Haigron
Publié dans: 2024, ISBN 9780443138966
Éditeur: Academia Press, Elsevier
DOI: 10.5281/ZENODO.12666190

Peer reviewed articles (1)

When Federated Learning Meets Watermarking: A Comprehensive Overview of Techniques for Intellectual Property Protection

Auteurs: Mohammed Lansari; Vincent Thouvenot; Katarzyna Kapusta; Gouenou Coatrieux; Reda Bellafqira; Olivier Bettan
Publié dans: Machine Learning & knowledge extraction, Numéro 5/4, 2023, ISSN 2504-4990
Éditeur: MDPI
DOI: 10.3390/make5040070

Droits de propriété intellectuelle

Patent (1)

PATH COMPUTATION IN A COMMUNICATION NETWORK

Numéro de demande/publication: 20 22050969
Date: 2022-10-24
Demandeur(s): ERICSSON TELECOMUNICAZIONI SPA