What would it cost to deploy this AI system in our hospital or company?+
The project does not publish specific licensing or deployment costs. INCISIVE was funded as a Research and Innovation Action, so the technology was developed with public funding. Commercial terms would need to be negotiated with the consortium, led by Maggioli SPA (Italy). The system is designed as modular services (AI Toolbox, Data Analytics, User Services), which may allow flexible pricing.
Can this scale to handle our imaging volume across multiple sites?+
Yes — the system was specifically designed for multi-site operation. It was validated across 8 pilot sites in 5 countries (Cyprus, Greece, Italy, Serbia, Spain) using federated learning, meaning the AI trains across sites without moving patient data. The repository was built to handle 1 PB of imaging data from 8 data-providing partners.
What about IP and licensing — who owns this technology?+
INCISIVE was developed by a 28-partner consortium led by Maggioli SPA, a large Italian IT company. IP ownership typically follows EU Horizon 2020 grant rules, where each partner owns the results they generate. Licensing arrangements would need to be discussed with the coordinator and relevant partners who developed specific components.
Does this meet healthcare data regulations like GDPR?+
The project was explicitly designed for compliance with ethical, legal, and privacy demands. It uses federated learning so patient data never leaves the originating hospital. The pan-European repository includes blockchain-based mechanisms for secure data donation and sharing. Based on available project data, the system addresses GDPR requirements by design.
How long would implementation take?+
The project ran from October 2020 to March 2024 and delivered final versions of all major components: the AI Toolbox, the Pan-European Repository, and the integrated prototypes. Pilot validations ran for 1.5 years across 8 sites. Based on available project data, the technology has been validated but would require integration work specific to your environment.
Which cancer types does this actually cover?+
INCISIVE validated its AI models across 4 cancer types: breast, prostate, colorectal, and lung cancer. These were tested in dedicated validation studies at 8 pilot sites with at least 2,600 patients participating. The models handle various cancer imaging scans, biological data, and electronic health records.
Is there ongoing support or is this a finished research project?+
The project officially closed in March 2024. However, with 28 partners including 11 industry organizations and 7 SMEs, several consortium members are likely pursuing commercial exploitation. The final deliverables include a stand-alone pan-European repository system and production-ready AI services. Contact the coordinator for post-project support options.
