Description du projet
Logiciel d’apprentissage en essaim pour le dépistage du cancer par IA
L’intelligence artificielle (IA) est appelée à révolutionner les soins de santé en améliorant le diagnostic médical. Toutefois, cela nécessite le partage de vastes ensembles de données sur les patients, processus qui se heurte souvent à plusieurs obstacles. L’apprentissage en essaim (Swarm Learning, SL) constitue un moyen de développer l’IA médicale dans le respect de la vie privée, mais l’accès à cette technologie reste limité. Dans ce contexte, le projet ODELIA, financé par l’UE, entend élaborer la première structure logicielle libre pour le SL, qui permettra de rationaliser le développement de solutions Al applicables au dépistage du cancer. Outre la mise au point d’une application médicale utile pour le diagnostic du cancer du sein, le projet démontrera les avantages cliniques de la technologie SL en termes de développement accéléré, de performances accrues et de généralisation robuste. Cette technologie permettra de sauver des milliers de vies.
Objectif
ArtifArtificial Intelligence (AI) will revolutionize healthcare as its diagnostic performance approaches that of clinical experts. In particular, in cancer screening, AI helps patients to make better-informed decisions and reduce medical error. However, this requires large datasets whose collection faces severe practical, ethical and legal obstacles. These obstacles can be overcome with swarm learning (SL) where partners jointly train AI models without sharing any data. Yet, access to SL technology is seriously limited because no studies have implemented SL in a true multinational setup, no practically usable implementation of SL is available, researchers & healthcare providers have no experience with setting up SL networks and policymakers are currently unaware of the broader implications of SL. ODELIA will address & solve these issues: ODELIA will build the first open-source software framework for SL, providing an assembly line for the streamlined development of AI solutions. To serve as a blueprint for future SL-based AI systems, ODELIA partners collaborate as a swarm to develop the first clinically useful AI algorithm for the detection of breast cancer in magnetic resonance imaging (MRI). The size of ODELIA's distributed database will exceed all previous studies and ODELIA's AI models will reach expert-level performance for breast cancer screening. Thereby, ODELIA will not only deliver a useful medical application, but prove the clinical benefit of SL in terms of accelerated development, increased performance and robust generalizability to ultimately save thousands of lives of European patients. ODELIA's success will push partners to serve as nuclei for the exponential growth of the SL network and extend SL to a multitude of medical applications. Thus, patients, healthcare providers and citizens in Europe will be provided with a digital infrastructure that enables development of expert-level AI tools on big data without compromising data safety and data privacy.
Champ scientifique
- natural sciencescomputer and information sciencesartificial intelligence
- natural sciencescomputer and information sciencesdatabases
- natural sciencescomputer and information sciencesdata sciencebig data
- medical and health sciencesclinical medicineoncologybreast cancer
- engineering and technologymedical engineeringdiagnostic imagingmagnetic resonance imaging
Programme(s)
- HORIZON.2.1 - Health Main Programme
- HORIZON.2.1.6 - Health Care Systems
Régime de financement
HORIZON-RIA - HORIZON Research and Innovation ActionsCoordinateur
1010 Wien
Autriche