In August, the go-ahead was given and together with our project partners DHC Business Solutions and the German Research Center for Artificial Intelligence, we are working like the fire department to make post market surveillance for medical technology even more efficient and indirectly increase patient safety.
The aim of the project is to develop a new service platform that will simplify and make more reliable the market observation (post market surveillance) and risk assessment of medical devices. State-of-the-art technologies and methods of artificial intelligence (AI) will be used. The research project SmartVigilance is funded by the Federal Ministry of Science and Research.
Manufacturers of medical devices are subject to strict regulations regarding the safety of their products. Regulations concern both product approval and the phase after market launch. Companies are obliged to monitor their products on the market and in use as part of post market surveillance (PMS) and to take appropriate measures to eliminate possible defects or reduce risks for patients.
What is already common practice for pharmaceutical companies, is now coming to medical technology companies to a greater extent: A large number of different documents from numerous sources must be identified, analyzed and evaluated to increase product safety. This is the ideal field for intelligent, automated processes based on our many years of experience in using AI in the life sciences.
Kornel Marko (Managing Director Averbis)
The new Internet-based service platform, which can automatically record and analyze dangerous incidents and product defects reported to the regulatory authorities and report them to the manufacturers, relies on various methods from the field of artificial intelligence (AI), especially natural language processing (NLP), machine learning and deep learning.
With this mix of technologies and methods, publicly accessible databases can be searched automatically, such as the Manufacturer and User Facility Device Experience (MAUDE) database of the U.S. Food and Drug Administration (FDA), the PubMed database for scientific literature in the life sciences and biomedicine, and from 2022 the European database EUDAMED (European Database on Medical Devices). International standards and terminological resources – including the International Classification of Diseases (ICD10), the Medical Dictionary for Regulatory Activities (MedDRA), Medical Subject Headings (MeSH) or the product classification of the Food & Drug Administration (FDA) – are also taken into account.
In addition to the integration of different AI methods, we are also advancing the topic of “Explainable AI”. This is because the results of AI solutions not only have to be correct, but they must also be presented in a comprehensible and comprehensible manner.
Kornel Marko (Managing Director Averbis)
SmartVigilance is funded by the Federal Ministry of Education and Research (BMBF), FKZ 01|S20028A.