The project is funded by Horizon Europe, and coordinated by Charité Universitätsmedizin Berlin.
VALIDATE (Validation of a Trustworthy AI-based Prognostic Tool for Predicting Patient Outcome in Acute Stroke) aims to develop, test, and validate a prognostic tool for predicting outcomes in acute stroke patients using Machine Learning-enabled Artificial Intelligence. The project leverages pre-clinical evidence, existing models, and a prototype of a clinical decision support system (patent pending).
The focus is on improving individualized treatment stratification, patient outcomes, and cost-effective healthcare resource allocation. The project particularly targets acute ischemic stroke, where AI-based prognostic tools can potentially enhance treatment effectiveness by considering complex relationships among various prognostic variables.
The initiative emphasizes adherence to Trustworthy AI guidelines outlined by the EU, ensuring regulatory, ethical, and robust considerations in the development, testing, and validation processes. The consortium involved in the project combines technical and medical expertise in machine learning with clinical proficiency from leading stroke hospitals, including collaboration with patient organizations. The industry partner, IBM iX Aperto, contributes software and implementation expertise, facilitating the integration of the prognostic tool into clinical practice. The ultimate goal is to surpass the current clinical state-of-the-art in acute stroke treatment by making a significant and sustainable impact on the treatment of acute stroke that will improve patient survival and outcome. The results should bethe basis of the inclusion of AI-based prognostic and predictive approaches in the clinical guidelines for acute stroke treatment.
The project aims to share its experiences through standard operating procedures (SOPs) and involve regulatory bodies to ensure the highest quality of recommendations.
As partners in the consortium, SimulaMet has three distinct roles:
- Contribute to the creation of a clinical trustworthy AI framework.
- Focus on the operationalisation of low-level AI ethical requirements
- Develop and integrate explainable AI algorithms in a demonstrator app for patient outcome prediction from stroke.
Other partners in the project are: Empirica Gesellschaft Fur Kommunikations Und Technologieforschung MBH, Techonoligcal University Dublin, Universitätsklinikum Heidelberg, Hadassah Medical Organisation, Fundacio Hospital Universitari Vall D'Hebron - Institut De Recerca, IBM, iX - Aperto Gmbh, SAFE Stroke Alliance For Europe, NORA S.L.