The Medicine and Healthcare products Regulatory Agency (MHRA) has published 10 principles for medical devices using artificial intelligence or machine learning software.
It has developed the principles with the US Food and Drug Administration and Health Canada, and said they are intended to lay a foundation for good practice in the field.
They cover key elements of good machine learning practices, including having an in-depth understanding of a model’s intended integration into clinical workflow, and the desired benefits and associate patient risks.
The 10 principles are as follows:
- Multi-disciplinary expertise is leveraged through the total lifecycle.
- Good software engineering and security practices are implemented.
- Clinical study participants and datasets are representative of the intended patient population.
- Training datasets are independent of test sets.
- Selected reference datasets are based upon best available methods.
- Model design is tailored to the available data and reflects the intended use of the device.
- Focus is placed on the performance of the human-AI team.
- Testing demonstrates device performance during clinically relevant conditions.
- Users are provided clear, essential information.
- Deployed models are monitored for performance and retraining risks are managed.
MHRA said: “These guiding principles further identify areas where the International Medical Device Regulators Forum (IMDRF), international standards organisations and other collaborative bodies could work together to advance GMLP. Areas of collaboration include research; creating educational tools and resources; regulatory policies and regulatory guidelines; international harmonisation; and consensus standards.
“We know that strong international partnerships will be essential part of empowering the wider sector to advance responsible innovations. We look forward to our continued collaborative work and engagement with the FDA and Health Canada and wider international health partners in this area.”
Image from iStock, metamorworks