Aidence is one of the first in its field to have been granted the CE Mark for its lung nodule management assistant Veye Chest under the new EU Medical Device Regulation (MDR, 2017/745).
With the certification of Veye Chest under the MDR as a class IIb medical device (previously class IIa), Aidence sets a new regulatory standard for this type of devices on the market. Gaining regulatory clearance in this device class requires a strong risk management process, as well as clinical data and a solid post-market clinical follow-up process.
Aidence’s AI solution Veye Chest automatically detects, classifies and tracks the growth of pulmonary nodules on chest CT scans. Deployed across multiple sites in the EU and the UK for both routine clinical practice and screening, Veye Chest is currently analysing thousands of scans each week.
Mark-Jan Harte, CEO of Aidence, said: “We believe Veye Chest is a premier tool for both routine clinical practice and lung cancer screening, contributing to increased diagnostic accuracy. And every step of the way, we are keenly aware of our responsibility to patients by respecting and exceeding the regulatory requirements for clinical quality and safety. I applaud the Aidence team today not only for building a robust medical device but also for their expertise and hard work obtaining this essential certification, ahead of many others.”
Leon Doorn, Aidence Head of Regulatory Compliance, added:“Certification of Veye Chest as a class IIb device under the MDR demonstrates that this type of devices require high-level evidence prior to use in clinical practice, and need monitoring by regulatory authorities, to guarantee their safe and effective use. I am proud to work for Aidence, where we spend significant effort in clinical validation and risk management.”
The MDR, which came into force on 26 May 2017, requires all medical devices (of Class IIa and higher) to reapply with their notified bodies for CE marking. Due to the COVID-19 health emergency, the deadline has been extended by one year, to 26 May 2021.
Background: AI in EU regulatory frameworks
Standalone software has been considered and regulated as a medical device since 2009 when a revision to the Medical Device Directive (MDD) introduced explicit requirements for software validation. One year later, the first discussions started around introducing an updated regulatory framework, to better address regulatory gaps and technological developments in the medical device industry. This led to the Medical Device Regulation (MDR) coming into force in 2017, with, amongst others, increased emphasis on software and stricter obligations for device manufacturers.
MDR introduces a new classification rule for medical software (rule 11), which considers the results returned by the software and how medical practitioners use these in clinical practice. Based on potential risk, software is classified as class I, IIa, IIb, or class III. Whilst under MDD, only software which provided information allowing direct diagnosis, MDR widened the scope by requiring all software that provides any kind of diagnostic support to be certified as class IIa or higher. For a more detailed look into the changes brought about by the MDR, consult this article.
To obtain certification, AI vendors must determine the applicable classification of their medical device, in cooperation with their Notified Bodies (i.e. an external, independent organisation licensed to assess the conformity of certain products before market release). Subsequently, the AI company submits technical documentation, including the clinical evaluation report and the risk assessment for the device, for regulatory approval.
Only certified AI solutions can be used in clinical practice, in conformity with the intended use for which they were approved.