The guidance aims to lay a strong foundation for implementing PCCPs as a means of managing specific device changes that usually require regulatory approval before marketing.
The following guiding principles ensure that medical devices incorporating artificial intelligence and machine learning (MLMD) can be modified, updated, and improved safely, effectively, and promptly in response to new data.
|
1. |
Focused and Bounded: - Having established procedures for making safe modifications to the device within the predefined limits of the PCCP. |
|
2. |
Risk-based - Ensuring that changes, whether individual or cumulative, remain suitable for the device and its operating environment over time. |
|
3. |
Evidence-Based: - Guaranteeing the device's ongoing safety and effectiveness through a PCCP, using scientifically and clinically validated methods and metrics to evaluate device performance. |
|
4. |
Transparent - Ensuring that stakeholders remain well-informed about the device's performance, both before and after implementing changes. |
|
5. |
Total Product Lifecycle (TPLC) Perspective: - Enhancing the quality and integrity of a PCCP by consistently considering the viewpoints of all stakeholders and incorporating risk management practices throughout the entire product lifecycle. |
References:
Predetermined Change Control Plans for Machine Learning-Enabled Medical Devices: Guiding Principles