Imagine your medical records are scattered across different doctors' offices in various formats, like handwritten notes and digital spreadsheets. This tool acts like a smart digital librarian that gathers all those messy pieces and organizes them into a clean, standardized folder. It helps patients easily share their health history with researchers or new doctors without needing a technical expert to translate the data.
Healthcare data is trapped in fragmented, unstructured formats that are not AI-ready. This creates a massive manual workload for data stewards and prevents the efficient use of personal health data for personalized medicine.
An AI-powered virtual assistant prototype featuring a backend library of curation tools and a frontend for human-AI interaction. Key deliverables include a Global Data Sharing Standard and a Data Quality Framework.
If you are a data intermediary dealing with fragmented patient records — this project developed an AI-powered virtual assistant that automates the cleaning and publishing of heterogeneous health data. This reduces the manual effort needed to make data AI-ready and reuse-ready.
If you are a hospital dealing with high workloads for clinical data stewards — this project developed AI-based curation tools that automate quality enhancement. This decreases the manual workload of staff managing breast cancer registries and cardiovascular records.
If you are a research organization dealing with unstructured narrative medical content — this project developed deep learning and NLP models for information extraction. This allows for faster creation of high-quality, interoperable datasets for personalized medicine research.
Based on available project data, no specific commercial pricing or cost model is mentioned; the project was funded by a EUR 7,720,615 EU contribution.
The project aims for scale across institutions at a national and EU level, specifically targeting the delivery of the European Health Data Space.
Based on available project data, specific IP or licensing agreements are not detailed, though it involves a consortium of 16 partners including 9 SMEs.
The system is designed for automation of quality enhancement and FAIRification in compliance with EU data privacy standards.
Integration is achieved through a data source onboarding tool that maps attributes from sources to specific workflows or target ontologies.
The consortium is highly commercially oriented, with a 50% industry ratio consisting of 8 industry partners, 9 of whom are SMEs. This balance between 3 universities and 1 research center suggests a strong push toward practical application and market viability rather than pure academic research, spanning 11 different countries.
Contact Universiteit Maastricht regarding the AIDAVA prototype
Contact us to connect with the 9 SMEs in the AIDAVA consortium for licensing opportunities.
