Open Consortium for Decentralized Medical Artificial Intelligence

Informations projet

ODELIA

N° de convention de subvention: 101057091

DOI

10.3030/101057091

Date de signature de la CE 17 Novembre 2022

Date de début 1 Janvier 2023

Date de fin 31 Decembre 2027

Financé au titre de

• Health

Coût total

€ 8 691 755,75

Contribution de l’UE

€ 8 691 755,00

Coordonné par

EIBIR GEMEINNUTZIGE GMBH ZUR FORDERUNG DER ERFORSCHUNG DER BIOMEDIZINISCHEN BILDGEBUNG
Austria

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

Websites, patent fillings, videos etc. (5)

Source code for the web front-end

The complete source code for the web front-end, shared publicly and open access

Project Video

First short video/animation to raise awareness of the ODELIA project

Updated web front-end

Updated web front-end (version 2.0) connected to a back-end AI-processing pipeline (can display the results of all ODELIA AI models or uploaded models on a fixed set of data or data uploaded by the user)

Web front-end

A web front-end with basic functionalities (version 1.0, can display the saved results of an AI model for a fixed set of data) will be operational and publicly usable by any stakeholder, including citizen scientists

Manual on setting up SL onsite

Manual describing how to set up SL onsite

Other (1)

Trained neural networks for detection of breast cancer in MRI images (including models trained on local data only and models trained in the swarm)

Trained neural networks for detection of breast cancer in MRI images. This includes trained neural network models trained on local data only and models trained in the swarm.

Documents, reports (8)

Report comparing the performance of baseline AI model against the winners of the competition, trained locally and in SL

Report comparing the performance of baseline AI model against the winners of the competition. This included neural networks trained locally and in SL.

Communication kit including website

Communication kit for ODELIA, including communication, dissemination and design assets such as a presentation template, standard slides introducing the project, a short project description, the logo and related visual assets, and the project website and online presence.

Report on summer schools

Report summarising the different summer schools organised by ODELIA

Midterm recruitment report

This report is due when 50% of the study population is recruited. The report includes an overview of the number of recruited participants by clinical sites, any problems in recruitment and, if applicable, a detailed description of implemented and planned measures to compensate for any incurred delays.

Guidance document on running SL in clinical environments

Guidance document for partners describing how to run SL in clinical environments

Instructions on training AI models with SL in a tangible demonstration case

Report with instructions on how to train AI models with SL based on a tangible demonstration case

Report comparing the performance of locally trained models against swarm-trained models in breast cancer screening
Study initiation package

Study initiation package (before enrolment of the first study participant)This deliverables includes:- Registration number of the clinical study in a registry meeting WHO Registry criteria- Final version of study protocol as approved by the regulator(s) / ethics committee(s)- Regulatory and ethics approvals required for the enrolment of the first study participant

Publications

Book chapters (1)

meval: A Statistical Toolbox for Fine-Grained Model Performance Analysis

Auteurs: Dishantkumar Sutariya, Eike Petersen
Publié dans: Lecture Notes in Computer Science, Fairness of AI in Medical Imaging, 2025
Éditeur: Springer Nature Switzerland
DOI: 10.1007/978-3-032-05870-6_19

Peer reviewed articles (10)

Swarm learning with weak supervision enables automatic breast cancer detection in magnetic resonance imaging

Auteurs: Oliver Lester Saldanha; Jiefu Zhu; Gustav Müller-Franzes; Zunamys I. Carrero; Nicholas R. Payne; Lorena Escudero Sánchez; Paul Christophe Varoutas; Sreenath Kyathanahally; Narmin Ghaffari Laleh; Kevin Pfeiffer; Marta Ligero; Jakob Behner; Kamarul A. Abdullah; Georgios Apostolakos; Chrysafoula Kolofousi; Antri Kleanthous; Michail Kalogeropoulos; Cristina Rossi; Sylwia Nowakowska; Alexandra Athanasiou; Raquel Perez-Lopez; Ritse Mann; Wouter Veldhuis; Julia Camps; Volkmar Schulz; Markus Wenzel; Sergey Morozov; Alexander Ciritsis; Christiane Kuhl; Fiona J. Gilbert; Daniel Truhn; Jakob Nikolas Kather
Publié dans: Communications Medicine, Numéro 5, 38 (2025), 2025, ISSN 2730-664X
Éditeur: Springer Nature
DOI: 10.18154/RWTH-2025-10188

Large language models could make natural language again the universal interface of healthcare

Auteurs: Jakob Nikolas Kather; Dyke Ferber; Isabella C. Wiest; Stephen Gilbert; Daniel Truhn
Publié dans: Nature Medicine, 2024, ISSN 1546-170X
Éditeur: Springer Nature
DOI: 10.1038/S41591-024-03199-W

Overcoming regulatory barriers to the implementation of AI agents in healthcare

Auteurs: Oscar Freyer; Sanddhya Jayabalan; Jakob N. Kather; Stephen Gilbert
Publié dans: Nature Medicine, 2025, ISSN 1546-170X
Éditeur: Springer Nature
DOI: 10.1038/S41591-025-03841-1

Scientific Reports

Auteurs: Müller-Franzes, Gustav; Khader, Firas; Siepmann, Robert; Han, Tianyu; Kather, Jakob Nikolas; Nebelung, Sven; Truhn, Daniel
Publié dans: Scientific Reports, Numéro (2025) 15:23979, 2025, ISSN 2045-2322
Éditeur: Springer Nature
DOI: 10.18154/RWTH-2025-06979

Reconstruction of patient-specific confounders in AI-based radiologic image interpretation using generative pretraining

Auteurs: Tianyu Han; Laura Zigutyte; Luisa Huck; Marc Huppertz; Robert Siepmann; Yossi Gandelsman; Christian Blüthgen; Firas Khader; Christiane Kuhl; Sven Nebelung; Jakob Nikolas Kather; Daniel Truhn
Publié dans: Cell Reports Medicine, 2024, ISSN 2666-3791
Éditeur: Cell Press (an imprint of Elsevier)
DOI: 10.48550/ARXIV.2309.17123

Nature Communications

Auteurs: Soroosh Tayebi Arasteh; Tianyu Han; Mahshad Lotfinia; Christiane Kuhl; Jakob Nikolas Kather; Daniel Truhn; Sven Nebelung
Publié dans: Nature Communications, Vol 15, Iss 1, Pp 1-12 (2024), 2024, ISSN 2041-1723
Éditeur: Nature Publishing Group
DOI: 10.48550/arxiv.2308.14120

Preserving fairness and diagnostic accuracy in private large-scale AI models for medical imaging

Auteurs: Soroosh Tayebi Arasteh; Alexander Ziller; Christiane Kuhl; Marcus Makowski; Sven Nebelung; Rickmer Braren; Daniel Rueckert; Daniel Truhn; Georgios Kaissis
Publié dans: Communications Medicine, 2024, ISSN 2730-664X
Éditeur: Springer Nature
DOI: 10.1038/S43856-024-00462-6

Scientific Reports

Auteurs: Firas Khader; Gustav Müller-Franzes; Soroosh Tayebi Arasteh; Tianyu Han; Christoph Haarburger; Maximilian Schulze-Hagen; Philipp Schad; Sandy Engelhardt; Bettina Baeßler; Sebastian Foersch; Johannes Stegmaier; Christiane Kuhl; Sven Nebelung; Jakob Nikolas Kather; Daniel Truhn
Publié dans: Scientific Reports, Vol 13, Iss 1, Pp 1-12 (2023), 2023, ISSN 2045-2322
Éditeur: Nature Publishing Group
DOI: 10.18154/rwth-2023-05893

Diffusion probabilistic versus generative adversarial models to reduce contrast agent dose in breast MRI

Auteurs: Gustav Müller-Franzes; Luisa Huck; Maike Bode; Sven Nebelung; Christiane Kuhl; Daniel Truhn; Teresa Lemainque
Publié dans: European Radiology Experimental, 2024, ISSN 2509-9280
Éditeur: SpringerOpen
DOI: 10.1186/S41747-024-00451-3

Nature Communications

Auteurs: Jan Clusmann; Dyke Ferber; Isabella C. Wiest; Carolin V. Schneider; Titus J. Brinker; Sebastian Foersch; Daniel Truhn; Jakob Nikolas Kather
Publié dans: Nature Communications, 2025, ISSN 2041-1723
Éditeur: Springer Nature
DOI: 10.1038/S41467-024-55631-X

Other (1)

A European Multi-Center Breast Cancer MRI Dataset

Auteurs: Gustav Müller-Franzes, Lorena Escudero Sánchez, Nicholas Payne, Alexandra Athanasiou, Michael Kalogeropoulos, Aitor Lopez, Alfredo Miguel Soro Busto, Julia Camps Herrero, Nika Rasoolzadeh, Tianyu Zhang, Ritse Mann, Debora Jutz, Maike Bode, et al
Publié dans: arXiv, Numéro 2506.00474, ISSN 2331-8422
Éditeur: Cornell University