Seminal
The magazine and atlas of EU-funded researchA new publication from the web and AI team at Scott Charnley

Atlas / Germany

European Innovation Council · Running

EUcanAI European Unified Cancer diagnostics and treatment AI agent

Coordinator
Ruprecht-Karls-Universitaet Heidelberg, Germany
Runs
1 September 2026 to 31 August 2030
Status
Running
EU funding
€3.9m
Total cost
—
Partners
14 organisations in 5 countries
Type
European Innovation Council
Grant agreement
101306931

What the project is doing

In the project's own words, from its summary on CORDIS.

The EUcanAI project pioneers a new paradigm for clinical management of central nervous system (CNS) cancers by creating a modular, agentic Generative AI ecosystem that unifies diagnosis and treatment into an integrated continuum of care. Current workflows in neuro-oncology are fragmented, slow, and highly dependent on individual expertise, leading to variable outcomes. EUcanAI transforms this landscape by developing interoperable AI agents that interact across three technological domains: (i) multimodal integration through an orchestration agent capable of synthesising clinical, imaging, histopathological and molecular data into transparent tumour board recommendations; (ii) medical data augmentation via generative models that produce synthetic MRI sequences, FFPE-quality histology from frozen sections, and rare diagnostic scenarios for model training and benchmarking; and (iii) knowledge representation and reasoning frameworks aligned with WHO CNS tumour classifications, ensuring transparent provenance, reproducibility, and explainable decision support.

The consortium unites 13 leading institutions across 3 EU Member States (Germany, Austria, Denmark), 1 Associated Partner (Korea), and 1 international collaborator (WHO-IARC), ensuring regional diversity, complementary expertise, and global external validation. This partnership enables access to the largest harmonised CNS tumour datasets in Europe (>10,000 cases) and world-class clinical infrastructures for multi-centre validation.

Through staged retrospective and prospective studies, EUcanAI will deliver clinically validated, human-in-the-loop AI systems supporting every step of CNS cancer care—from patient history extraction to intraoperative diagnostics and therapeutic planning. By embedding regulatory compliance with the EU AI Act, MDR and GDPR from inception, EUcanAI ensures trustworthy, sovereign and clinically adoptable AI.

Topics

Who is involved

OrganisationCityCountryRole
Ruprecht-Karls-Universitaet HeidelbergHeidelbergGermanyCoordinator
Medizinische Universitaet WienWienAustriaParticipant
Medizinische Universitat InnsbruckInnsbruckAustriaParticipant
Deutsches Krebsforschungszentrum HeidelbergHeidelbergGermanyParticipant
Johann Wolfgang Goethe-Universitaet Frankfurt Am MainFrankfurt Am MainGermanyParticipant
Technische Hochschule KolnKolnGermanyParticipant
Universitaet GreifswaldGreifswaldGermanyParticipant
Universitaetsklinikum EssenEssenGermanyParticipant
Universitatsklinikum BonnBonnGermanyParticipant
Universitatsklinikum ErlangenErlangenGermanyParticipant
Kobenhavns UniversitetKobenhavnDenmarkParticipant
Centre International De Recherche Sur Le CancerLyon 7emeFranceParticipant
Asan Medical CenterSeoulSouth KoreaAssociatedpartner
Yonsei UniversitySeoulSouth KoreaAssociatedpartner

Source: CORDIS, European Commission. Data of 5 Nov 2026, reused under the Commission's reuse policy. View on CORDIS · Grant DOI