REALM
Real-world-data Enabled Assessment for heaLth regulatory decision-Making
Duration
2023-2026
Coordinated by
Universiteit Maastricht
Projects Partners
13
Project funding
6.6M €
The Project
REALM is developing a decentralised, secure evaluation and certification framework for medical device software, using blockchain technology and intelligent contracts, offering transparency, repeatability, and traceability while fully complying with General Date Protection Regulation.
Through AI-powered regulatory tools, the R&I project evaluates algorithms and models in terms of bias, privacy, accuracy, legal accountability, ethics, etc., which improves data-driven decision-making.
REALM facilitates access to real-world data from public repositories as well as generation of synthetic data. Leverage them for testing medical software algorithms, for scenarios with limited available samples.
Overall, REALM promotes trust and adoptability of the machine learning solutions through explainable AI approaches that address transparency, accountability, explainability concerns and potential biases induced by model predictions.
Our role in the project
Within the REALM project, VPH plays a central role in connecting scientific innovation with the wider stakeholder community. We lead the stakeholder mapping process and act as a liaison with regulatory actors to ensure the project’s outcomes align with real-world needs. VPH also contributes to the REALM Academy by preparing targeted training materials and designing effective engagement strategies through workshops and webinars. In parallel, we drive the dissemination and communication of REALM’s results, ensuring that its impact reaches key audiences across research, regulation, and industry.
Outcomes
Regulatory AI-sandbox for medical device software (REALM-AI):
REALM-AI sandbox platform is about building a federated infrastructure that includes real-world and synthetic data, along with an advanced regulatory toolbox to evaluate high-risk AI-based medical device software. Technical and ELSI experts within the REALM consortium, shape the sandbox environment that offer versatile features for benchmarking, pre-market assessment and post-market surveillance of medical device software.
REALM Academy: is the educational hub of the REALM project, providing targeted training and resources for regulators, notified bodies, competent bodies, HTA agencies, and software developers working in the evaluation, certification, and oversight of Medical Device Software (MDSW). Through webinars, online training, and workshops, REALM Academy ensures that stakeholders across Europe are equipped with the knowledge, tools, and methodologies to make data-driven, transparent, and standardized decisions in the rapidly evolving landscape of AI-based medical software.
