Group Research and Development

DNV webinar series on AI, medical devices, MDR and the upcoming AI Act

DNV’s hybrid webinar series on AI, medical devices, Regulation (EU) 2017/745 on Medical Devices (MDR) and the upcoming AI Act (AIA). The webinars aim to highlight some relevant aspects of the conformity requirements for medical device software which use AI. The webinars focus primarily on MDR requirements, but touch upon requirements from upcoming AI Act in parallel.

There is not a single day without breaking news regarding the disruptive power of artificial intelligence (AI) from generative language processing to image recognition,  financial analysis, or even cultural content creation. In medicine, AI carries the promise of better and more personalized medicine, harvesting on the growing amount of  digital health data. However, AI, like any other medical device, needs to comply to the basic principle of medical practice, “do no harm”. In Europe, this translates into the  obligation of compliance to the Medical Device Regulation and other relevant laws in the Union. This series of webinars aims to highlight some key aspects of conformity management of medical device software and AI, which are prerequisites to the AI in MDSW - Ai mind logoapplication of AI medical devices in clinical practice. As the regulatory landscape evolves with new technologies, this series also aims to provide insights into the upcoming AI regulation in Europe that manufacturers or developers of AI software may want to consider early on in their development process.  

The  webinars are co-organised with the AI-Mind consortium of which DNV is a partner. AI-Mind brings together 15 partners from 8 countries to develop AI-based tools for dementia risk prediction of people with mild cognitive impairment. The project has received funding from the European Union’s Horizon 2020 research and innovation AI in MDSW - eu flagprogramme under grant agreement No 964220. https://www.ai-mind.eu/ . To join the AI-Mind mailing list, please visit the subscription page https://www.ai-mind.eu/subscribe/

The webinars are targeted at AI-Mind members and stakeholders, the IVS network, and other relevant stakeholders researching and using AI in healthcare including  clinicians, researchers, lawyers, hospital management and IT support. 

The webinars will be delivered by medical device software expert, Dr Frédéric Courivaud from DNV’s Healthcare Research Programme. 


Agenda

Webinar Series

AI & Medical Device Software; Views on specific conformity assessment aspects in the context of the current European regulatory landscape

AI carries the promise to improve healthcare delivery and personal medicine. At the same time this new technology must meet the same regulatory requirements as any other medical device prior to its clinical use. This webinar will highlight in particular the potential automation bias and the general requirements for safety and performance.

ML in medical device software; A Notified Body’s perspective in a changing regulatory landscape

This second webinar focuses on the regulatory requirements regarding clinical data and clinical evaluation under the MDR.

Anticipating the AIA for the conformity assessment of medical device Software

This third and last webinar focuses on the interplay between MDR and upcoming AI Act regarding the conformity management of AI medical device software.

Disclaimer

The considerations and general views expressed in this webinar series are intended as a resource in the management of regulatory requirements of AI MDSW in Europe and do not represent professional or legal advice of any kind. These considerations and views shall be revised in the light of upcoming regulatory updates and the publication of new guidance documents, such as from the medical device coordination group (MDCG). Placing on the market or putting AI/ML medical device software into service in the European Economic Area (EEA) requires compliance with MDR. DNV encourages manufacturers developing AI/ML MDSW for medical use to seek regulatory knowledge and support early in the design phase.