Thanks to the advanced use of Natural Language Processing techniques, the D-Hygea Lab team, part of the Department of Electronics, Information and Bioengineering – Politecnico di Milano, has developed a tool capable of classifying and aggregating into a local database safety notices related to medical devices available on both European and non-European markets.

The research, coordinated by PhD student Yijun Ren, is part of the European project CORE-MD (COordinating REsearch and Evidence for Medical Devices) and was recently published in the prestigious journal Nature Digital Medicine.

Safety notices are issued by manufacturers to national competent authorities when malfunctions are identified in marketed devices. Having an automated tool like the one developed by the D-Hygea Lab makes it possible to identify recurring patterns, analyse temporal trends in malfunctions, and compare products or companies within the same category. This not only helps to detect potential issues in advance but also strengthens market surveillance by regulatory authorities and notified bodies.

In addition to supporting post-market surveillance, the system enables manufacturing companies to conduct proactive research and gather valuable information to meet the periodic surveillance requirements set out by the new European Medical Device Regulation.